Re: Web development with Python 3.1
mario ruggier wrote: With respect to to original question regarding web frameworks + database and Python 3, all the following have been available for Python 3 since the day Python 3.0 was released: QP, a Web Framework http://pypi.python.org/pypi/qp/ Durus, a Python Object Database (the "default" in qp, for user sessions, etc) http://pypi.python.org/pypi/Durus/ Evoque, state-of-the-art templating engine http://pypi.python.org/pypi/evoque/ (this one is available for py3.0 since a little later, 21-jan-2009) All the above also runs on python 2.4 (thru to python 3) For the record, you may see the list of all pypi packages availabe for Python 3 at: http://pypi.python.org/pypi?:action=browse&show=all&c=533 m. Thanks for those links Mario - I'll check them out asap.. Regards, Alan -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Oct 30, 6:23 pm, Dotan Cohen wrote: > The point is that I want to use only _Python_ features, not > Django/Mako/whatever features. Pure python has a builtin templating system -- its called % See http://simonwillison.net/2003/Jul/28/simpleTemplates/ -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
With respect to to original question regarding web frameworks + database and Python 3, all the following have been available for Python 3 since the day Python 3.0 was released: QP, a Web Framework http://pypi.python.org/pypi/qp/ Durus, a Python Object Database (the "default" in qp, for user sessions, etc) http://pypi.python.org/pypi/Durus/ Evoque, state-of-the-art templating engine http://pypi.python.org/pypi/evoque/ (this one is available for py3.0 since a little later, 21-jan-2009) All the above also runs on python 2.4 (thru to python 3) For the record, you may see the list of all pypi packages availabe for Python 3 at: http://pypi.python.org/pypi?:action=browse&show=all&c=533 m. -- http://mail.python.org/mailman/listinfo/python-list
Re. Web development with Python 3.1
Rober Kern wrote > But if you insist, you may be interested in Breve: > http://pypi.python.org/pypi/Breve/ Thanks for that! Viva internal DSLs! [Sorry -- cut my teeth on lisp] Is there anything like this for xml? Well I guess that is a slightly wrong (if not straight stupid) question. Maybe something along this line: Takes a spec as DTD or Relax-ng or somethn and produces something like Breve? -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> notmm uses Python 2.6 and will probably work just fine with Python >> 3000. >> The only reference to "notmm" that I could find in Google was this thread! > "I am free, no matter what rules surround me. If I find them > tolerable, I tolerate them; if I find them too obnoxious, I break > them. I am free because I know that I alone am morally responsible for > everything I do." -- Robert A. Heinlein > Heinlein said that? It's been a long time since I've read Heinlein, and his quotes are as vivid as his books. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Oct 30, 7:01 pm, erob wrote: > On Oct 28, 5:16 am, "Diez B. Roggisch" wrote: > > > > > Dotan Cohen schrieb: > > > >> While I know that to be true in the general sense, from what I've > > >> looked at Django and other frameworks it seems that the web frameworks > > >> push the coder to use templates, not letting him near the HTML. > > > >> For instance, I was looking for a class / framework that provided a > > >> proven method of decoding cookies (setting them is no problem), > > >> decoding POST and GET variables, escaping variables for safe entry > > >> into MySQL, and other things. Django and the other frameworks seem to > > >> force the user to use templates. I just want the functions, and to > > >> print the HTML as stdout to the browser making the request. I had to > > >> settle on PHP to do this, which admittedly is what PHP was invented to > > >> do. However, for obvious reasons, I would have prefered to code in > > >> Python. In fact, I still would. > > > > I should probably expand on this: > > > > How can I get an array with all the GET variables in Python? > > > How can I get an array with all the POST variables in Python? > > > How can I get an array with all the COOKIE variables in Python? > > > How can I get the request URI path (everything after > > >http://[www.?]example.com/)? > > > > That's all I want: no templates and nothing between me and the HTML. > > > The HTTP headers I can output to stdout myself as well. > > > Again: if you insist on doing everything yourself - then of course any > > library or framework isn't for you. > > > But then do you deal with headers correctly? Do you respect character > > encodings? Form-encodings? Is your generated HTML valid? Are > > timestamp-formats generated according to RFCs for your cookies? Do you > > parse content negotiation headers? > > > I think you underestimate the task it is to make a webapplication good. > > And even if not, what you will do is ... code your own webframework. > > Because there is a lot of boilerplate otherwis. If that's a > > learning-experience your after, fine. > > > Besides, yes, you can get all these things nonetheless. You just don't > > need them most of the time. > > > And at least pylons/TG2 lets you return whatever you want instead, as a > > string. Not via "print" though - which is simply only for CGI, and no > > other means (e.g. mod_wsgi) of python-web-programming. > > > Diez > > notmm uses Python 2.6 and will probably work just fine with Python > 3000. > > Cheers, > > Etienne > > P.S - We all don't think in the same box. "I am free, no matter what rules surround me. If I find them tolerable, I tolerate them; if I find them too obnoxious, I break them. I am free because I know that I alone am morally responsible for everything I do." -- Robert A. Heinlein -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Oct 28, 5:16 am, "Diez B. Roggisch" wrote: > Dotan Cohen schrieb: > > > > >> While I know that to be true in the general sense, from what I've > >> looked at Django and other frameworks it seems that the web frameworks > >> push the coder to use templates, not letting him near the HTML. > > >> For instance, I was looking for a class / framework that provided a > >> proven method of decoding cookies (setting them is no problem), > >> decoding POST and GET variables, escaping variables for safe entry > >> into MySQL, and other things. Django and the other frameworks seem to > >> force the user to use templates. I just want the functions, and to > >> print the HTML as stdout to the browser making the request. I had to > >> settle on PHP to do this, which admittedly is what PHP was invented to > >> do. However, for obvious reasons, I would have prefered to code in > >> Python. In fact, I still would. > > > I should probably expand on this: > > > How can I get an array with all the GET variables in Python? > > How can I get an array with all the POST variables in Python? > > How can I get an array with all the COOKIE variables in Python? > > How can I get the request URI path (everything after > >http://[www.?]example.com/)? > > > That's all I want: no templates and nothing between me and the HTML. > > The HTTP headers I can output to stdout myself as well. > > Again: if you insist on doing everything yourself - then of course any > library or framework isn't for you. > > But then do you deal with headers correctly? Do you respect character > encodings? Form-encodings? Is your generated HTML valid? Are > timestamp-formats generated according to RFCs for your cookies? Do you > parse content negotiation headers? > > I think you underestimate the task it is to make a webapplication good. > And even if not, what you will do is ... code your own webframework. > Because there is a lot of boilerplate otherwis. If that's a > learning-experience your after, fine. > > Besides, yes, you can get all these things nonetheless. You just don't > need them most of the time. > > And at least pylons/TG2 lets you return whatever you want instead, as a > string. Not via "print" though - which is simply only for CGI, and no > other means (e.g. mod_wsgi) of python-web-programming. > > Diez notmm uses Python 2.6 and will probably work just fine with Python 3000. Cheers, Etienne P.S - We all don't think in the same box. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On 2009-10-30 15:55 PM, Dotan Cohen wrote: It is clear and obvious. But it has the "template engine" duplicating a function that Python has built in. My goal is to learn reusable Python (reusable for non-web projects). My goal is not to find the quickest way to a website. Please correct me if I'm wrong, but you did learn HTML, CSS and SQL, didn't you ? How do any of these languages relates to Python ?-) HTML and CSS yes, but just enough SQL to fake it. None of those languages relate to Python, each language performs a different function. However, the template language looks completely redundant as it duplicates functionality of the language that I intend to learn. Templating languages serve a different, highly specialized purpose. They may have similar flow constructs, but the context and specializations matter a lot. Use the right language for the job. Sometimes the job is a weird mix of concerns and requires a weird mix of constructs to be convenient. If you're writing web apps, you should learn a templating language. If you're primarily concerned with learning Python and nothing else, you should find a different kind of project. But if you insist, you may be interested in Breve: http://pypi.python.org/pypi/Breve/ -- Robert Kern "I have come to believe that the whole world is an enigma, a harmless enigma that is made terrible by our own mad attempt to interpret it as though it had an underlying truth." -- Umberto Eco -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> I took a look a both yesterday. They are both generic text templating
> systems that seem to pretty much do the same thing. I suspect you will
> prefer Mako since it avoids duplicating Python's comtrol structures. But I
> think it worthwhile to look at both anyway since doing so will help to
> separate the concepts from the particular implementations.
>
> My take on them is this: when text mixes code that is meant to be
> interpreted and text data meant to be taken literally, some means must be
> devised to distinguish the two. In programs files, the code is left unquoted
> and the text data is quoted. In template files, the marking is reversed: the
> literal text is left unquoted and the code *is* quoted. In Mako, expressions
> are quoted with braces ({...}), single statements with '%' prefix, and
> multiple statements as well as Mako tags with <% ...>.
>
Thanks, Terry, that should save me some time.
--
Dotan Cohen
http://what-is-what.com
http://gibberish.co.il
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> It is clear and obvious. But it has the "template engine" duplicating >> a function that Python has built in. My goal is to learn reusable >> Python (reusable for non-web projects). My goal is not to find the >> quickest way to a website. > > Please correct me if I'm wrong, but you did learn HTML, CSS and SQL, didn't > you ? How do any of these languages relates to Python ?-) > HTML and CSS yes, but just enough SQL to fake it. None of those languages relate to Python, each language performs a different function. However, the template language looks completely redundant as it duplicates functionality of the language that I intend to learn. >> won't help me learn Python any more than learning PHP would. >> So going from PHP to Django is pointless. > > Please get out of the "server page" mindset. Templates are just that : > templates. Period. You don't use templates to write your *applicative* code. > I understand that. >> I meant reusable knowledge, not reusable code. > > There's very few "reusable knowledge" to be gained from templating code > anyway- even if using a "python-based" template system like Mako. Unless you > think doing a loop or a conditional are things you need to learn ?-) > I would say that one needs to know how to do a for loop in Python, in order to program in Python! >>> Then use Mako - it uses plain Python to manage the presentation logic. >> >> Now we're talking! I will look further into Pylons and Mako. > > Beware that the above comments about how less there's to learn from the > templates code still apply - basically, the "programming" part of a template > is restricted to simple branching, iterations, and variable substitution. > I see. >> I don't expect to ever have a "team", > > Possibly not. But as strange as it migh be, there are other peoples that do, > and most of these frameworks are written by professional web programmers. > I therefore counter that these frameworks were designed for people unlike myself, which easily explains my resistance to them. >> The point is that I want to use only _Python_ features, not >> Django/Mako/whatever features. > > If so, you shouldn't use *any* third-part libs, and possibly not even the > stdlib. > THat is going just a bit far! > More seriously : reinventing the wheel - unless for educational purposes - > is not really pythonic. If what you want is to learn, write your own > framework, template system, form handling etc... You'll very certainly find > out that they suck big time compared to existing projects, but you'll learn > _at least_ one thing: that writing a sensible non-trivial framework or lib > is *hard*. > Actually, I konw just how hard it is. PHP and some C in university. >> However I am aware that some things I >> should not touch for security reasons. That is why I want a framework: >> to provide the security aspects of things like converting UTF-8, > > from what and to what ?-) > Into and out of the database. > string / unicode encoding and decoding is a builtin Python feature. > I know, but I need to play with it a bit before I become comfortable with how it is handled. >> database escaping, > > Already provided by any DB-API compliant connector, at least if correctly > used. > > Now there's *much* more in security (specially when doing web programming) > than this... Of course. But these are the relevant issues for Python. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen wrote:
It is clear and obvious. But it has the "template engine" duplicating
a function that Python has built in.
...
Then use Mako - it uses plain Python to manage the presentation logic. And
if you go for Mako, then you might as well switch to Pylons. Great framework
too (better than Django on some parts FWIW), but you'll probably need much
more time to be fully productive with it (power and flexibility come with a
price...).
Now we're talking! I will look further into Pylons and Mako.
I took a look a both yesterday. They are both generic text templating
systems that seem to pretty much do the same thing. I suspect you will
prefer Mako since it avoids duplicating Python's comtrol structures. But
I think it worthwhile to look at both anyway since doing so will help to
separate the concepts from the particular implementations.
My take on them is this: when text mixes code that is meant to be
interpreted and text data meant to be taken literally, some means must
be devised to distinguish the two. In programs files, the code is left
unquoted and the text data is quoted. In template files, the marking is
reversed: the literal text is left unquoted and the code *is* quoted. In
Mako, expressions are quoted with braces ({...}), single statements with
'%' prefix, and multiple statements as well as Mako tags with <% ...>.
Terry Jan Reedy
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit :
Look at this "templating code":
{% for entry in blog_entries %}
{{ entry.title }}
{{ entry.body }}
{% endfor %}
What's the problem ? Simple, clear and obvious.
It is clear and obvious. But it has the "template engine" duplicating
a function that Python has built in. My goal is to learn reusable
Python (reusable for non-web projects). My goal is not to find the
quickest way to a website.
Please correct me if I'm wrong, but you did learn HTML, CSS and SQL,
didn't you ? How do any of these languages relates to Python ?-)
Why would I want to learn that when Python already has a real for
loop ?
If you know even 101 Python, understanding the above code shouldn't be such
a great challenge !-)
It is not a great challenge, but it is redundant. Learning Django
language
Django's templating language, actually
won't help me learn Python any more than learning PHP would.
So going from PHP to Django is pointless.
Please get out of the "server page" mindset. Templates are just that :
templates. Period. You don't use templates to write your *applicative* code.
I know HTML, and I have a goal of learning Python for it's
reusability (desktop apps, for instance).
Using templates instead of harcoding HTML in the applicative code actually
plays a big part when it comes to reusability.
I meant reusable knowledge, not reusable code.
There's very few "reusable knowledge" to be gained from templating code
anyway- even if using a "python-based" template system like Mako. Unless
you think doing a loop or a conditional are things you need to learn ?-)
Telling Django's template
loader where to look for templates is just a matter of configuration, so
using templates you can segment your whole project in small, possibly
reusable apps - then in another project where you need the same features but
with a totally different presentation, it's just a matter of writing
project-specific templates. Quite a few django apps work that way, and they
really save you a LOT of time. This wouldn't be possible using your beloved
antipattern...
I think that the time that I invest codng in Python as opposed to
Django Template Language is time well spent.
Then write your own templating system - and your own framework FWIW !-)
I don't want to learn some
"templating language" that duplicates what Python already has built
in!
Then use Mako - it uses plain Python to manage the presentation logic.
Now we're talking! I will look further into Pylons and Mako.
Beware that the above comments about how less there's to learn from the
templates code still apply - basically, the "programming" part of a
template is restricted to simple branching, iterations, and variable
substitution.
Now wrt/ "why having a distinct templating language", there are pros and
cons. On the pros side, having a language that is clearly restricted to
presentation logic prevents you (and anyone else working on the project -
and sometimes you have to deal with well-below-average guys in your team) to
put anything else than presentation logic in the templates.
I don't expect to ever have a "team",
Possibly not. But as strange as it migh be, there are other peoples that
do, and most of these frameworks are written by professional web
programmers.
Now, as far as I'm concerned, having Mako instead of Django's templates
wouldn't bother me at all. But Django has it's own template system, which
from experience get the job done (believe me, there are way worse
alternatives), and the overall qualities and features of the whole framework
are IMHO enough to justify learning it's templating language.
I see. I will have to spend some time with each to decide, though.
That's usually the best thing to do - take a couple days doing the
tutorials, and choose the one that fits your brain.
Given the restricted and rather unintersting nature of pure presentation
logic, you won't learn much from this part of the project anyway. You'd
probably learn way more using Django's templates and learning how to write
your own template tags. Or if you prefer Mako / Pylons, learning to make
proper use of Mako's advanced features. But one thing is clear - if you
persist on applying your favorite antipattern, you'll just waste more of
your time. Your choice, as usual...
The point is that I want to use only _Python_ features, not
Django/Mako/whatever features.
If so, you shouldn't use *any* third-part libs, and possibly not even
the stdlib.
More seriously : reinventing the wheel - unless for educational purposes
- is not really pythonic. If what you want is to learn, write your own
framework, template system, form handling etc... You'll very certainly
find out that they suck big time compared to existing projects, but
you'll learn _at least_ one thing: that writing a sensible non-trivial
framework or lib is *hard*.
Once again, been here, done that...
However I am aware that some things I
should not touc
Re: Web development with Python 3.1
>> Look at this "templating code":
>> {% for entry in blog_entries %}
>> {{ entry.title }}
>> {{ entry.body }}
>> {% endfor %}
>
> What's the problem ? Simple, clear and obvious.
>
It is clear and obvious. But it has the "template engine" duplicating
a function that Python has built in. My goal is to learn reusable
Python (reusable for non-web projects). My goal is not to find the
quickest way to a website.
>> Why would I want to learn that when Python already has a real for
>> loop ?
>
> If you know even 101 Python, understanding the above code shouldn't be such
> a great challenge !-)
>
It is not a great challenge, but it is redundant. Learning Django
language won't help me learn Python any more than learning PHP would.
So going from PHP to Django is pointless.
>> I know HTML, and I have a goal of learning Python for it's
>> reusability (desktop apps, for instance).
>
> Using templates instead of harcoding HTML in the applicative code actually
> plays a big part when it comes to reusability.
I meant reusable knowledge, not reusable code.
> Telling Django's template
> loader where to look for templates is just a matter of configuration, so
> using templates you can segment your whole project in small, possibly
> reusable apps - then in another project where you need the same features but
> with a totally different presentation, it's just a matter of writing
> project-specific templates. Quite a few django apps work that way, and they
> really save you a LOT of time. This wouldn't be possible using your beloved
> antipattern...
>
I think that the time that I invest codng in Python as opposed to
Django Template Language is time well spent.
>> I don't want to learn some
>> "templating language" that duplicates what Python already has built
>> in!
>
> Then use Mako - it uses plain Python to manage the presentation logic. And
> if you go for Mako, then you might as well switch to Pylons. Great framework
> too (better than Django on some parts FWIW), but you'll probably need much
> more time to be fully productive with it (power and flexibility come with a
> price...).
>
Now we're talking! I will look further into Pylons and Mako.
> Now wrt/ "why having a distinct templating language", there are pros and
> cons. On the pros side, having a language that is clearly restricted to
> presentation logic prevents you (and anyone else working on the project -
> and sometimes you have to deal with well-below-average guys in your team) to
> put anything else than presentation logic in the templates.
>
I don't expect to ever have a "team", but as a hobbiest and not a
coder I myself am "below average". I promise you that _will_improve_,
though.
> Bless me for I have sinned - sometimes, when you're tired and under pressure
> to deliver in time, it's tempting to add a Q&D hack in the presentation part
> instead of DoingTheRightThing(tm). This can save you some "precious" minutes
> now. The problem is that usually your boss won't give you the resources to
> clean up this mess once you delivered, and next time you have to work on
> this project - fix or evolution - you have to deal with this hack. Then with
> another, yet another, until your code is nothing more than an unmaintainable
> heap of junk. Been here, done that, bought the T-shirt :(
>
> Also remember that most of the time, a web app is a team effort, and even if
> *you* don't fail to temptation, others may do so. And finally, the team
> usually involve "HTML coders" - who, despite what the name seems to imply,
> are usually skilled enough to handle the presentation logic by themselves,
> so you the application programmer can concentrate on applicative code. From
> experience, they usually prefer a simple straightforward - even if somehow
> restricted - templating language.
>
> Now, as far as I'm concerned, having Mako instead of Django's templates
> wouldn't bother me at all. But Django has it's own template system, which
> from experience get the job done (believe me, there are way worse
> alternatives), and the overall qualities and features of the whole framework
> are IMHO enough to justify learning it's templating language.
>
I see. I will have to spend some time with each to decide, though.
>> I think that I will use the HTTP-abstraction features of Django, but
>> disregard the templates. My code might be uglier, but the knowledge
>> will be transferable to other projects. My ultimate goal is not to
>> make the latest greatest website. My ultimate goal is to port my
>> perfectly functional website to a language that will benefit me by
>> knowing it.
>
> Given the restricted and rather unintersting nature of pure presentation
> logic, you won't learn much from this part of the project anyway. You'd
> probably learn way more using Django's templates and learning how to write
> your own template tags. Or if you prefer Mako / Pylons, learning to make
> proper use of Mako's advanced features. But one thing is clear - if you
> persist on a
Re: Web development with Python 3.1
Diez B. Roggisch a écrit : The point is that using templates allows you to express your rendering-logic in terms of the desired output language (HTML in this case). Well, for Django, Mako, Cheetah and quite a few others, this might not be _that_ true - you can use any of the templating systems to generate almost any kind of text, not just HTML, and the templating logic itself is _not_ expressed "in terms of the desired output language" (ie: the "template langage" elements are not valid HTML). And the overall separation of rendering from programming is a *great* thing for maintainability. +1 on this - even if some may argue that this shouldn't prevent you from using Python to handle the presentation logic (whitespace problems set aside, and cf Mako and Cheetah). -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen wrote:
>> Perhaps this might better answer your questions:
>> http://docs.djangoproject.com/en/dev/topics/templates/#id1
>>
>
> Actually, that example just proves my point! Look at this "templating
> code":
> {% for entry in blog_entries %}
> {{ entry.title }}
> {{ entry.body }}
> {% endfor %}
>
> Why would I want to learn that when Python already has a real for
> loop? I know HTML, and I have a goal of learning Python for it's
> reusability (desktop apps, for instance). I don't want to learn some
> "templating language" that duplicates what Python already has built
> in!
The point is that using templates allows you to express your rendering-logic
in terms of the desired output language (HTML in this case).
This becomes even more apparent in genshi/KID, which use XML-namespaces
inside normal XHTML-documents to express their logic. And thus you can
- use a decent XML/HTML-editor to work with them, syntax-hilighting and
everthing included.
- communicate better with people who are only used to work with HTML
- aren't a slave to Python's whitespace rules which are fine for Python,
but not for HTML.
- validate the HTML and be sure you don't f*ck it up with forgotten
print-statements
And the overall separation of rendering from programming is a *great* thing
for maintainability.
Diez
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Bruno Desthuilliers wrote: Dotan Cohen a écrit : I don't want to learn some "templating language" that duplicates what Python already has built in! Then use Mako - it uses plain Python to manage the presentation logic. And if you go for Mako, then you might as well switch to Pylons. Great framework too (better than Django on some parts FWIW), but you'll probably need much more time to be fully productive with it (power and flexibility come with a price...). Good to know about Mako -- I don't want to have to learn yet another language, either. Not, at least, until I have achieved Python Mastery! :) Guess I better find some time to read through my Pylons book... ~Ethan~ -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit :
Perhaps this might better answer your questions:
http://docs.djangoproject.com/en/dev/topics/templates/#id1
Actually, that example just proves my point!
Which one ?
Look at this "templating code":
{% for entry in blog_entries %}
{{ entry.title }}
{{ entry.body }}
{% endfor %}
What's the problem ? Simple, clear and obvious.
Why would I want to learn that when Python already has a real for
loop ?
If you know even 101 Python, understanding the above code shouldn't be
such a great challenge !-)
I know HTML, and I have a goal of learning Python for it's
reusability (desktop apps, for instance).
Using templates instead of harcoding HTML in the applicative code
actually plays a big part when it comes to reusability. Telling Django's
template loader where to look for templates is just a matter of
configuration, so using templates you can segment your whole project in
small, possibly reusable apps - then in another project where you need
the same features but with a totally different presentation, it's just a
matter of writing project-specific templates. Quite a few django apps
work that way, and they really save you a LOT of time. This wouldn't be
possible using your beloved antipattern...
I don't want to learn some
"templating language" that duplicates what Python already has built
in!
Then use Mako - it uses plain Python to manage the presentation logic.
And if you go for Mako, then you might as well switch to Pylons. Great
framework too (better than Django on some parts FWIW), but you'll
probably need much more time to be fully productive with it (power and
flexibility come with a price...).
Now wrt/ "why having a distinct templating language", there are pros and
cons. On the pros side, having a language that is clearly restricted to
presentation logic prevents you (and anyone else working on the project
- and sometimes you have to deal with well-below-average guys in your
team) to put anything else than presentation logic in the templates.
Bless me for I have sinned - sometimes, when you're tired and under
pressure to deliver in time, it's tempting to add a Q&D hack in the
presentation part instead of DoingTheRightThing(tm). This can save you
some "precious" minutes now. The problem is that usually your boss won't
give you the resources to clean up this mess once you delivered, and
next time you have to work on this project - fix or evolution - you have
to deal with this hack. Then with another, yet another, until your code
is nothing more than an unmaintainable heap of junk. Been here, done
that, bought the T-shirt :(
Also remember that most of the time, a web app is a team effort, and
even if *you* don't fail to temptation, others may do so. And finally,
the team usually involve "HTML coders" - who, despite what the name
seems to imply, are usually skilled enough to handle the presentation
logic by themselves, so you the application programmer can concentrate
on applicative code. From experience, they usually prefer a simple
straightforward - even if somehow restricted - templating language.
Now, as far as I'm concerned, having Mako instead of Django's templates
wouldn't bother me at all. But Django has it's own template system,
which from experience get the job done (believe me, there are way worse
alternatives), and the overall qualities and features of the whole
framework are IMHO enough to justify learning it's templating language.
I think that I will use the HTTP-abstraction features of Django, but
disregard the templates. My code might be uglier, but the knowledge
will be transferable to other projects. My ultimate goal is not to
make the latest greatest website. My ultimate goal is to port my
perfectly functional website to a language that will benefit me by
knowing it.
Given the restricted and rather unintersting nature of pure presentation
logic, you won't learn much from this part of the project anyway. You'd
probably learn way more using Django's templates and learning how to
write your own template tags. Or if you prefer Mako / Pylons, learning
to make proper use of Mako's advanced features. But one thing is clear -
if you persist on applying your favorite antipattern, you'll just waste
more of your time. Your choice, as usual...
My 2 cents...
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : 2009/10/29 Albert Hopkins : On Wed, 2009-10-28 at 16:38 +0200, Dotan Cohen wrote: return HttpResponse(unmaintanable_html % data) That's fine for single variables, but if I need to output a table of unknown rows? I assume that return means the end of the script. Therefore I should shove the whole table into a variable and then copy that variable to the array "data"? No, if you use a templating system like Django's then basically you pass a QuerySet to your template. A QuerySet is basically a pointer to a SQL query, for example. The templating system just knows to expect an iterable, it can be a list of rows or it can be a QuerySet which does a fetch from a database. No need to shove an entire table into a variable. Sounds like the system is trying to outsmart the programmer here. Why so ??? -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit :
Ok, here's a Django example without a template:
# views.py
def index(request):
torvalds = request.GET.get("torvalds", "No torvalds here")
return HttpResponse("%s" % torvalds)
Allright, but what if I need to output in the middle of the page? Say,
one row of a table that is neither the end nor beginning of the code?
I ask because the construct "return" makes me believe that this is not
good for this purpose.
A django "view" function returns a full HttpResponse. It's nothing like
PHP's "server page" scheme - here you first build your full response,
then pass it back to the framework (which will deal with the frontal
webserver).
and here's an exemple using a template:
(snip)
Again, how could I output something in the middle of the page with the
> templates?
Already answered elsewhere in this post.
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> Perhaps this might better answer your questions:
> http://docs.djangoproject.com/en/dev/topics/templates/#id1
>
Actually, that example just proves my point! Look at this "templating code":
{% for entry in blog_entries %}
{{ entry.title }}
{{ entry.body }}
{% endfor %}
Why would I want to learn that when Python already has a real for
loop? I know HTML, and I have a goal of learning Python for it's
reusability (desktop apps, for instance). I don't want to learn some
"templating language" that duplicates what Python already has built
in!
I think that I will use the HTTP-abstraction features of Django, but
disregard the templates. My code might be uglier, but the knowledge
will be transferable to other projects. My ultimate goal is not to
make the latest greatest website. My ultimate goal is to port my
perfectly functional website to a language that will benefit me by
knowing it.
--
Dotan Cohen
http://what-is-what.com
http://gibberish.co.il
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : What do you mean by "in the middle of the page"? Do you mean, for instance, the behavior of "middle.php" in the following PHP example: Is that what you are after? Yes, that is what I am after. Django's templating system has an "include" statement, but also something you don't have in PHP : an "extends" statement. Yes, template inheritance. This means that you can define the general basic layout for all your site in one "base" template, and make your other templates just fill the "blanks" you defined in the base template. FWIW, I'm using Django has an example but quite a few other templating systems have this "template inheritance" feature one way or another... For instance, if one were to look at the source code of http://dotancohen.com they would see "". All the HTML up to that point was output by bigginin.inc.php. Similarly, near the end exists "", all the code from there is generated by end.inc.php. These two files are included in every page of the site. And ? Do you really thing we'd all be wasting time using templating systems too dumb to do even such a simple thing ??? Perhaps this might better answer your questions: http://docs.djangoproject.com/en/dev/topics/templates/#id1 -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> As is, in my case. Actually, what use case is there for having Apache >> reprocess the HTML output of the script? > > compression, cache, whatever... > Thanks. I actually did think of compression. >>> It's not that it was unclear, but that it's innaccurate. "outputting to >>> stdout" is an implementation detail, and should not be exposed at the >>> applicative code level. Dealing with appropriate abstraction - here, an >>> HttpResponse object - is far better (well, IMHO of course... - standard >>> disclaimers, YMMV etc). >>> >> >> I see. I believe that is called Dotan's Razor: a slight inaccuracy >> saves a lengthy explanation. > > but also impacts your "mental map" of what's going on... You're thinking in > terms of streams and stdout, which, while not that far from actual > implementation (in the end it always boils down to bits sent over the > wires...), might not be the right level of abstraction when it comes to > application programming. > Seeing how the templating engines work, I now know why my choice of words bothered you. In fact, I was thinking in terms of PHP's linear progress through the code which is not quite valid here. Thank you for taking the time to set me straight. > What I saw is already complex enough to justify using better tools IMHO. Or > to make better use of the one you have. > I am beginning to see how true that is. > I'd personnaly suggest either Mako (possibly one of the best Python > templating systems) or Django's (well documented and beginner-friendly). > I will examine them both this weekend. Thanks! -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : (snip) Yes, I like to separate HTML from code. However, often I need to output something in the middle of the page. How could I do that with a template? Perhaps learning to use some templating system could help ?-) I'd personnaly suggest either Mako (possibly one of the best Python templating systems) or Django's (well documented and beginner-friendly). -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : Clearly. I was referring to stdout being send to the browser as the http response. s/browser/web server/ - it's the web server that reads your app's stdout and send it (as is or not FWIW) back to the client. As is, in my case. Actually, what use case is there for having Apache reprocess the HTML output of the script? compression, cache, whatever... And this "output to stdout" thingie is just the ipc scheme used for CGI - there are other possible interfaces between the application code and the web server. Other possible interfaces between the application code and the web server? Depends on the web server, obviously. With Apache you have things like mod_python, mod_wsgi, or mod_php if you go that route. FWIW, while I would not necessarily advise using mod_python (unless of course you really need deep Apache integration instead of independance from the frontal web server), reading the mod_python's manual might teach you a lot about the processing of a HTTP request and what you can do... > I think I mentioned that, but I apologize for being unclear. It's not that it was unclear, but that it's innaccurate. "outputting to stdout" is an implementation detail, and should not be exposed at the applicative code level. Dealing with appropriate abstraction - here, an HttpResponse object - is far better (well, IMHO of course... - standard disclaimers, YMMV etc). I see. I believe that is called Dotan's Razor: a slight inaccuracy saves a lengthy explanation. but also impacts your "mental map" of what's going on... You're thinking in terms of streams and stdout, which, while not that far from actual implementation (in the end it always boils down to bits sent over the wires...), might not be the right level of abstraction when it comes to application programming. Sorry, but all I can "replace" here are the header and footer - if I want to generate a different markup for the "content here" part, I have to modify your applicative code. I've written web apps that way myself (some 7 years ago), and still have to maintain some web apps written that way, you know... Quite so, I though that is what you wanted. Yes, the HTML is hard-coded into the script. I am learning to abstract and even use object-oriented approaches, though. You can have efficient abstraction and decoupling without resorting to OO. Also, remember that PHP is first a templating language... I google, but cannot find any exapmles of this online. Well, no one in it's own mind would still use CGI (except perhaps for very trivial stuff) when you have way better solutions. What I am doing _is_ trivial. What I saw is already complex enough to justify using better tools IMHO. Or to make better use of the one you have. I understand that your goal is to help me. I am a hard learner: I recognize that you are teaching me the better way, but I need convincing as to why it is better. "Bruno said so" is good, It isn't. but "saves you coding time down the line" is a lot more convincing! Well, my POV is obviously biased - I've been doing mostly web programming for 6 or 7 years now, and not as a hobby. When you have several new projects to ship within a short timeline while still maintaining older projects, you really start thinking hard about readability, maintainability, and possible reuse. But wrt/ hard-coded embedded HTML vs templating (even in it's simpler form), it's not a matter of "bruno said so" - it's a well-known antipattern. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
2009/10/29 Albert Hopkins : > On Wed, 2009-10-28 at 16:38 +0200, Dotan Cohen wrote: >> > return HttpResponse(unmaintanable_html % data) >> > >> >> That's fine for single variables, but if I need to output a table of >> unknown rows? I assume that return means the end of the script. >> Therefore I should shove the whole table into a variable and then copy >> that variable to the array "data"? >> > No, if you use a templating system like Django's then basically you pass > a QuerySet to your template. A QuerySet is basically a pointer to a SQL > query, for example. The templating system just knows to expect an > iterable, it can be a list of rows or it can be a QuerySet which does a > fetch from a database. No need to shove an entire table into a > variable. Sounds like the system is trying to outsmart the programmer here. > Another advantage if templates for many is that it allows programmers to > do what they do best (write code) while letting web designers do what > they do best (designing web pages) without them walking over each other > (that much). > I do see that advantage. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> I think you're misunderstanding how Python and the web work together. Python > is not PHP -- it was not designed to replace HTML, and it is in itself not a > templating language. If you would like to use Python code embedded in HTML, > the way PHP is typically used, you may want to look at Python Server Pages > (http://www.modpython.org/live/current/doc-html/pyapi-psp.html) which are > provided by mod_python. > I see, thank you! This is what has been bothering and confusing me. I started with HTML and added PHP later, and the PHP fit in nicely embedded in the HTML, and even though it's now HTML embedded in PHP the principals are the same. Python doesn't seem to work that way, and I have been having a hard time wrapping my head around just why. > Just so you know (as I think this is what is causing much of the > misunderstanding here), most Python web frameworks *do not* subscribe to the > traditional model of the web, where URLs represent files on a server. (In > the case of PHP, Python Server Pages, or CGI scripts, these are augmented > files which are executed by the server, but it's the same concept). Instead, > they typically map URLs to arbitrary code. > I don't think that I like that idea. I will give it a fair chance, though, now that it problem has been identified. Thank you very much for helping me understand this important distinction. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Martin v. Löwis wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 (I can hear many of you shouting "don't - start with 2.6"), but as far as I can see, none of the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are Python3 compatible yet. That's not entirely true, see http://wiki.python.org/moin/PortingDjangoTo3k Regards, Martin Interesting link. Thanks for the info. Regards, Alan -- http://mail.python.org/mailman/listinfo/python-list
Re: Re: Web development with Python 3.1
John Nagle wrote:
Alan Harris-Reid wrote:
I am very much new to Python, and one of my first projects is a simple
data-based website. I am starting with Python 3.1
Until MySQLdb gets ported to something later than Python 2.5, support
for a "data-based web site" probably has to be in Python 2.5 or earlier.
The C module situation for Python 3.x still isn't very good.
Realistically, the production version of Python is 2.5. This process
is taking long enough that I'm worried that Python 3.x could do for
Python what Perl 6 did for Perl - provide an upgrade path that nobody
takes.
HTMLTemplate
("http://py-templates.sourceforge.net/htmltemplate/index.html";)
is a minimal templating system for fill-in-the-blanks template work in
Python.
Actually, if you have HTMLTemplate, FCGI, and MySQLdb, you have enough to
do a back-end database.
There are some advantages to libraries (you call them) over
"frameworks"
(they call you) if you're doing something unusual. Frameworks are more
useful if you're doing yet another "Web 2.0" web site.
John Nagle
Thanks for the advice John. I am concerned at what you say about the
uptake of 3.x - you could be right (although I hope you eventually
turn-out to be wrong :-) ).
Regards
Alan
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Wed, 28 Oct 2009 07:08:12 -, John Nagle wrote: Alan Harris-Reid wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 Until MySQLdb gets ported to something later than Python 2.5, support for a "data-based web site" probably has to be in Python 2.5 or earlier. Aside from the deprecation warning, MySQLdb appears to work perfectly well on my Ubuntu Python 2.6. Since my spare-time teach-myself activity has been playing with it, I'd have expected to notice lack of support by now. -- Rhodri James *-* Wildebeest Herder to the Masses -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Wed, 2009-10-28 at 16:38 +0200, Dotan Cohen wrote: > > return HttpResponse(unmaintanable_html % data) > > > > That's fine for single variables, but if I need to output a table of > unknown rows? I assume that return means the end of the script. > Therefore I should shove the whole table into a variable and then copy > that variable to the array "data"? > No, if you use a templating system like Django's then basically you pass a QuerySet to your template. A QuerySet is basically a pointer to a SQL query, for example. The templating system just knows to expect an iterable, it can be a list of rows or it can be a QuerySet which does a fetch from a database. No need to shove an entire table into a variable. > > > - second solution: do basically the same thing with a template > system - > > which will give you much more power and options... > > > > I will look further into the Django templates, I promise. But I would > still like to know how to work with Python proper. Another advantage if templates for many is that it allows programmers to do what they do best (write code) while letting web designers do what they do best (designing web pages) without them walking over each other (that much). -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Wed, 28 Oct 2009 14:15:54 -0700, Dotan Cohen wrote: What do you mean by "in the middle of the page"? Do you mean, for instance, the behavior of "middle.php" in the following PHP example: Is that what you are after? Yes, that is what I am after. For instance, if one were to look at the source code of http://dotancohen.com they would see "". All the HTML up to that point was output by bigginin.inc.php. Similarly, near the end exists "", all the code from there is generated by end.inc.php. These two files are included in every page of the site. I think you're misunderstanding how Python and the web work together. Python is not PHP -- it was not designed to replace HTML, and it is in itself not a templating language. If you would like to use Python code embedded in HTML, the way PHP is typically used, you may want to look at Python Server Pages (http://www.modpython.org/live/current/doc-html/pyapi-psp.html) which are provided by mod_python. Just so you know (as I think this is what is causing much of the misunderstanding here), most Python web frameworks *do not* subscribe to the traditional model of the web, where URLs represent files on a server. (In the case of PHP, Python Server Pages, or CGI scripts, these are augmented files which are executed by the server, but it's the same concept). Instead, they typically map URLs to arbitrary code. -- Rami Chowdhury "Never attribute to malice that which can be attributed to stupidity" -- Hanlon's Razor 408-597-7068 (US) / 07875-841-046 (UK) / 0189-245544 (BD) -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Wed, 2009-10-28 at 15:32 +0200, Dotan Cohen wrote: > > def index(request): > >unmaintanable_html = """ > > > > > >Index > > > > > >Embedded HTML is a PITA > >but some like pains... > > > > > > """ > >return HttpResponse(unmaintanable_html) > > > > And if I need to add a variable or three in there? Static HTML I can > do without Python. > Put the variables in a dict and then return unmaintanable_html % varDict Or, if you want the the static HTML in a file then you can return open(static_html_filename).read() % varDict Add any complexity beyond that though and you're pretty much on your way to writing a template engine ;-) -a -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> What do you mean by "in the middle of the page"? Do you mean, for instance,
> the behavior of "middle.php" in the following PHP example:
>
>
> include_once("beginning.inc.php");
>
> include_once("middle.php");
>
> include_once("end.inc.php");
>
> ?>
>
> Is that what you are after?
>
Yes, that is what I am after. For instance, if one were to look at the
source code of http://dotancohen.com they would see "". All the HTML up to that point was output by bigginin.inc.php.
Similarly, near the end exists "", all the
code from there is generated by end.inc.php. These two files are
included in every page of the site.
--
Dotan Cohen
http://what-is-what.com
http://gibberish.co.il
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Wed, 28 Oct 2009 12:42:17 -0700, Dotan Cohen wrote: I've already given you that for TG: class RootController(BaseController): @expose() def page(self, torwalds=None): return "%s" % (torwalds if torwalds else "" Does return mean that this could not be used in the middle of a page? [snip] Yes, I like to separate HTML from code. However, often I need to output something in the middle of the page. How could I do that with a template? What do you mean by "in the middle of the page"? Do you mean, for instance, the behavior of "middle.php" in the following PHP example: Is that what you are after? -- Rami Chowdhury "Never attribute to malice that which can be attributed to stupidity" -- Hanlon's Razor 408-597-7068 (US) / 07875-841-046 (UK) / 0189-245544 (BD) -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> Actually, currently in the example url >> http://example.com/path/to/script.py?var1=hello&var2=world the script >> is /home/user/site-name/public_html/path/ (with no filename extension) >> and the script.py is actually page.html which is an arbitrary, >> descriptive string and not part of the script filename. > > Better than the first example wrt/ SEO, but this still ties your "url space" > to implementation AFAICT. > Yes, but for the purposes of my needs this is fine and preferable. SEO was in fact the reason that I started doing this. >> Clearly. I was referring to stdout being send to the browser as the >> http response. > > s/browser/web server/ - it's the web server that reads your app's stdout and > send it (as is or not FWIW) back to the client. As is, in my case. Actually, what use case is there for having Apache reprocess the HTML output of the script? > And this "output to stdout" > thingie is just the ipc scheme used for CGI - there are other possible > interfaces between the application code and the web server. > Other possible interfaces between the application code and the web server? Maybe you are referring to the interface with the database server? Enlighten me, please! I am not a programmer by trade, only by hobby. >> I think I mentioned that, but I apologize for being >> unclear. > > It's not that it was unclear, but that it's innaccurate. "outputting to > stdout" is an implementation detail, and should not be exposed at the > applicative code level. Dealing with appropriate abstraction - here, an > HttpResponse object - is far better (well, IMHO of course... - standard > disclaimers, YMMV etc). > I see. I believe that is called Dotan's Razor: a slight inaccuracy saves a lengthy explanation. > Sorry, but all I can "replace" here are the header and footer - if I want to > generate a different markup for the "content here" part, I have to modify > your applicative code. I've written web apps that way myself (some 7 years > ago), and still have to maintain some web apps written that way, you know... > Quite so, I though that is what you wanted. Yes, the HTML is hard-coded into the script. I am learning to abstract and even use object-oriented approaches, though. >> Does it? I will look into that. I assume that basic support means >> making the cookie, GET and POST variables easily accessible. > > GET and POST at least - I don't exactly remember how it works for cookies. > That much I should be able to google when I get to it. >> I google, >> but cannot find any exapmles of this online. > > Well, no one in it's own mind would still use CGI (except perhaps for very > trivial stuff) when you have way better solutions. > What I am doing _is_ trivial. However, I thank you for explaining my options and I will take a second look at them. >> Hey! Since when am I not under attack? :) > > Mmm... yes, I tend to be a bit offensive sometimes - please don't take it > personnaly and forgive me for my lack of social skills. What I express are > of course _my_ opinions (most of them based on experience but that's not the > point), and I've never killed anyone for having a different POV - even if > I'm pretty sure they are wrong !-) > Be offensive! I take no offence, I understand that your goal is to help me. I am a hard learner: I recognize that you are teaching me the better way, but I need convincing as to why it is better. "Bruno said so" is good, but "saves you coding time down the line" is a lot more convincing! -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> I've already given you that for TG:
>
> class RootController(BaseController):
>
> @expose()
> def page(self, torwalds=None):
> return "%s" % (torwalds if torwalds
> else ""
>
Does return mean that this could not be used in the middle of a page?
> Of course nobody would do it that way. Instead, you'd define a template
>
> xmlns:py='http://genshi.edgewall.org/'>${torvalds}
>
>
> And then your code becomes:
>
>
>
> @expose("your.template.path")
> def page(self, torwalds=None):
> return dict(torwalds=torwalds)
>
>
> Separation of code from HTML-code. Which most people agree is a good thing.
> And no worries about having to make torwalds a string.
>
Yes, I like to separate HTML from code. However, often I need to
output something in the middle of the page. How could I do that with a
template?
> Now let's say you want torwalds to be not-empty, and a number:
>
> @expose("your.template.path")
> @validate(dict(torwalds=Int(not_empty=True)),
> error_handler=some_error_handler_controller_action)
> def page(self, torwalds=None):
> return dict(torwalds=torwalds)
>
That is nice, I will read more about the error_handler functions. Thanks.
--
Dotan Cohen
http://what-is-what.com
http://gibberish.co.il
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> It
>> did not look like I had any control over the tags at all.
>
> I can tell you you have full control over the whole HTTP response's content
> and headers. And FWIW, this has nothing to do with templating systems.
>
That's good to know. Actually, it's more that good: I am looking into
Django again.
>> Can you provide a code example of printing the variable "torvalds" in
>> the GET request?
>> http://example.com/page.py?torvalds=tux
>> Should return this:
>> tux
>>
>> And something similar for a POST request?
>
>
> Using which framework ? If it's about Turbogears, I don't use it so I can't
> answer. If it's using Pylons, you can use almost whatever templating system
> you want - or not use any. If it's with Django, do you want it with or
> without the templating system ?
>
I'd love to see the non-templating Pylons example, and/or the Django
example. Better yet, a link to example code or the appropriate chapter
of TFM would be welcome. So far as I can tell, the fine manual does
not cover non-template output.
>> I hate to ask for code, but this conversation would be easier that way.
>
> Mmmm May I suggest that you first read the relevant documentation for
> the aforementionned frameworks ?
>
Yes, please! Where in the Django docs is the relevant chapter?
> Ok, here's a Django example without a template:
>
> # views.py
> def index(request):
> torvalds = request.GET.get("torvalds", "No torvalds here")
> return HttpResponse("%s" % torvalds)
>
>
Allright, but what if I need to output in the middle of the page? Say,
one row of a table that is neither the end nor beginning of the code?
I ask because the construct "return" makes me believe that this is not
good for this purpose.
> and here's an exemple using a template:
>
> # templates/index.html
>
>
>
> {{ torvalds }}
>
>
>
> # views.py
> def index(request):
> torvalds = request.GET.get("torvalds", "No torvalds here")
> return render_to_response("index.html", dict(torvalds=torvalds))
>
Again, how could I output something in the middle of the page with the
templates?
--
Dotan Cohen
http://what-is-what.com
http://gibberish.co.il
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Diez B. Roggisch wrote: John Nagle schrieb: Alan Harris-Reid wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 Until MySQLdb gets ported to something later than Python 2.5, support for a "data-based web site" probably has to be in Python 2.5 or earlier. The usual FUD from you. http://pypi.python.org/pypi/MySQL-python/1.2.3c1 And of course there are other alternatives to databases, which happily run under 2.6 as well. Diez The main web site for MySQLdb says it is only supported through Python 2.5. "http://sourceforge.net/projects/mysql-python/"; "MySQL support for Python. MySQL versions 3.23-5.1; and Python versions 2.3-2.5 are supported." There does seem to be a 2.6 download for Linux, but the SourceForge download for Windows is for Python 2.5. Maybe the 2.6 version, with a "c1" suffix, is an unsupported release candidate. John Nagle -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : *Why* on earth do you think using a templating system will give you any less control on the generated HTML ? Because I am wrong. I have already come to that conclusion. I know that Python is far enough developed that if I feel that I am fighting it, then _I_ am in the wrong, not Python. However, I still need to see that Python can do what I need in a comfortable fashion. And if I need to add a variable or three in there? - first solution : use string formatting python 2.x example: unmaintanable_html = """ %(title)s %(h1)s %(text)s """ data = dict( title="index", h1="Poor man's templating", text="Won't get you very far..." ) return HttpResponse(unmaintanable_html % data) That's fine for single variables, but if I need to output a table of unknown rows? What do you think templates are for ? I assume that return means the end of the script. It means that this function returns an HttpResponse object. Therefore I should shove the whole table into a variable and then copy that variable to the array "data"? If you really love feeling the pain, that's a possible solution. I for one would rather use a template... But once again, your choice !-) - second solution: do basically the same thing with a template system - which will give you much more power and options... I will look further into the Django templates, I promise. But I would still like to know how to work with Python proper. Using the appropriate tool for the job is "proper Python". The appropriate tool for generating formatted text with static and dynamic parts is a template system. Good news, there are quite a few available in Python. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> *Why* on earth do you think using a templating system will give you any less > control on the generated HTML ? > Because I am wrong. I have already come to that conclusion. I know that Python is far enough developed that if I feel that I am fighting it, then _I_ am in the wrong, not Python. However, I still need to see that Python can do what I need in a comfortable fashion. >> And if I need to add a variable or three in there? > > - first solution : use string formatting > > python 2.x example: > > unmaintanable_html = """ > > > %(title)s > > > %(h1)s > %(text)s > > > """ > > data = dict( > title="index", > h1="Poor man's templating", > text="Won't get you very far..." > ) > > return HttpResponse(unmaintanable_html % data) > That's fine for single variables, but if I need to output a table of unknown rows? I assume that return means the end of the script. Therefore I should shove the whole table into a variable and then copy that variable to the array "data"? > - second solution: do basically the same thing with a template system - > which will give you much more power and options... > I will look further into the Django templates, I promise. But I would still like to know how to work with Python proper. >> Static HTML I can >> do without Python. > > Sorry, I forgot to setup a database etc for this example !-) > I meant that I would like to see how to output variables, which you have shown. Thanks. I am stubborn, but I am willing to learn. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit :
>>> I insist on handling the HTML myself.
>> I just don't get how embedding HTML in applicative code - a well-known
>> antipattern FWIW - relates to "handling the HTML yourself". Can you
*please*
>> explain how a templating system prevents you from "handling the HTML"
>> yourself.
>>
>
>>From the little that I played with Django, it looked like I would (for
> the sake of argument I am just using the term output, but the
> intention should be clear) output "Hello, world!" and Django would
> mangle that into Hello, world!.
I don't know how you got that result - but it's certainly *not* what I
get here.
> It
> did not look like I had any control over the tags at all.
I can tell you you have full control over the whole HTTP response's
content and headers. And FWIW, this has nothing to do with templating
systems.
>
And
even if not, what you will do is ... code your own webframework.
>>> That is why I am looking for a class that handles the backend stuff,
>>> but lets _me_ handle the HTML.
>> For God's sake : *why on earth do you believe that using a
templating system
>> will make you loose _any_ control on the HTML code ???*
>>
>
> Because that's what I've seen of them.
I use Django's templates to generate any kind of textual content - not
only HTML. I could even use them to generate Python source code. And
that's true for quite a few other templating systems around (Mako,
Jinja, Cheetah, just to name a few).
>
And at least pylons/TG2 lets you return whatever you want instead,
as a
string. Not via "print" though - which is simply only for CGI, and no
other
means (e.g. mod_wsgi) of python-web-programming.
>>> I am trying to follow you here. What is "return whatever you want"?
>>> Return HTML to stdout to be sent to the browser?
>
>> please leave stdout out of here - it's HTTP, so what we have are HTTP
>> requests and HTTP responses - not stdin nor stdout. Using these
streams as a
>> mean of communication between the web server and the applicative code is
>> just plain old low-level GCI stuff. Your application code shouldn't
have any
>> knowledge of this implementation detail - it should receive (a suitable
>> representation of) an HTTP request, and generate (a suitable
representation
>> of) an HTTP response.
>
> Can you provide a code example of printing the variable "torvalds" in
> the GET request?
> http://example.com/page.py?torvalds=tux
> Should return this:
> tux
>
> And something similar for a POST request?
Using which framework ? If it's about Turbogears, I don't use it so I
can't answer. If it's using Pylons, you can use almost whatever
templating system you want - or not use any. If it's with Django, do you
want it with or without the templating system ?
> I hate to ask for code, but this conversation would be easier that way.
Mmmm May I suggest that you first read the relevant documentation
for the aforementionned frameworks ?
Ok, here's a Django example without a template:
# views.py
def index(request):
torvalds = request.GET.get("torvalds", "No torvalds here")
return HttpResponse("%s" % torvalds)
and here's an exemple using a template:
# templates/index.html
{{ torvalds }}
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : While I know that to be true in the general sense, from what I've looked at Django and other frameworks it seems that the web frameworks push the coder to use templates, not letting him near the HTML. Well... there must be a reason, for sure... No ? ֹYes, but I don't like it. Why so ? What's your problem with templating systems, exactly ? They don't give me the control over the HTML that I would like. I'm probably stuck in 1998 when they didn't exist, but I like to write the HTML myself. *Why* on earth do you think using a templating system will give you any less control on the generated HTML ? Django and the other frameworks seem to force the user to use templates. Not at all. Nothing prevents you from shooting yourself in the foot and generating HTML "by hand" in your view functions (or request handler methods etc, depending on the framework). How is this done in Django, then? def index(request): unmaintanable_html = """ Index Embedded HTML is a PITA but some like pains... """ return HttpResponse(unmaintanable_html) And if I need to add a variable or three in there? - first solution : use string formatting python 2.x example: unmaintanable_html = """ %(title)s %(h1)s %(text)s """ data = dict( title="index", h1="Poor man's templating", text="Won't get you very far..." ) return HttpResponse(unmaintanable_html % data) - second solution: do basically the same thing with a template system - which will give you much more power and options... Static HTML I can do without Python. Sorry, I forgot to setup a database etc for this example !-) -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : I have no idea what reverse url generation is. It's a mean to avoid hard-coding urls in your application code - the url is fully generated by the 'url mapping' system. I don't need that, see below. I assume that the user will call http://example.com/path/to/script.py?var1=hello&var2=world script.py would live in /home/user/site-name/public_html/path/to/ Now try to use apache url rewrites to use "nice urls" (like, "///" instead of "path/to/script-serving-articles.wtf?id=42"). Good luck. Actually, currently in the example url http://example.com/path/to/script.py?var1=hello&var2=world the script is /home/user/site-name/public_html/path/ (with no filename extension) and the script.py is actually page.html which is an arbitrary, descriptive string and not part of the script filename. Better than the first example wrt/ SEO, but this still ties your "url space" to implementation AFAICT. I would prefer to output everything from to with print statements. I don't want some framework wrapping my output in tags, I want the tags to be part of the output. You're confusing the framework and the templating system. The framework by itself won't wrap your "output" in anything, nor even forces you to "output" HTML. This is what I was hoping to hear. I don't remember the details, but I had a hard time outputting the HTML myself in Django. I will play with it again soon if I get the chance. Also FWIW, it's HTTP, so we don't have "outputs", we have HTTP responses. Clearly. I was referring to stdout being send to the browser as the http response. s/browser/web server/ - it's the web server that reads your app's stdout and send it (as is or not FWIW) back to the client. And this "output to stdout" thingie is just the ipc scheme used for CGI - there are other possible interfaces between the application code and the web server. I think I mentioned that, but I apologize for being unclear. It's not that it was unclear, but that it's innaccurate. "outputting to stdout" is an implementation detail, and should not be exposed at the applicative code level. Dealing with appropriate abstraction - here, an HttpResponse object - is far better (well, IMHO of course... - standard disclaimers, YMMV etc). Yes, why not? Let's say I want to reuse your app with a different presentation. Do I have to fork the whole application and send my HTML coder dig into your applicative code just to rewrite the HTML parts ? No, just replace two files. In that manner it is templating. Again, an example from the current PHP code: Sorry, but all I can "replace" here are the header and footer - if I want to generate a different markup for the "content here" part, I have to modify your applicative code. I've written web apps that way myself (some 7 years ago), and still have to maintain some web apps written that way, you know... Like I said before, I don't want to have to maintain the functions that turn the HTTP environment into Python variables, "HTTP environment" ??? Oh, you mean HTTP requests etc... Well, the cgi module already provides some basic support here. Does it? I will look into that. I assume that basic support means making the cookie, GET and POST variables easily accessible. GET and POST at least - I don't exactly remember how it works for cookies. I google, but cannot find any exapmles of this online. Well, no one in it's own mind would still use CGI (except perhaps for very trivial stuff) when you have way better solutions. or the code that manages database connections. Functions that escape data to be sent to MySQL (to prevent sql injection) would be nice. Any DB-API compliant module already provide this - granted you use it correctly. I know. This is will use correctly, I promise! Other than that, it's all on me. Your choice... Hey! Since when am I not under attack? :) Mmm... yes, I tend to be a bit offensive sometimes - please don't take it personnaly and forgive me for my lack of social skills. What I express are of course _my_ opinions (most of them based on experience but that's not the point), and I've never killed anyone for having a different POV - even if I'm pretty sure they are wrong !-) -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen wrote:
>>> I insist on handling the HTML myself.
>>
>> I just don't get how embedding HTML in applicative code - a well-known
>> antipattern FWIW - relates to "handling the HTML yourself". Can you
>> *please* explain how a templating system prevents you from "handling the
>> HTML" yourself.
>>
>
>>From the little that I played with Django, it looked like I would (for
> the sake of argument I am just using the term output, but the
> intention should be clear) output "Hello, world!" and Django would
> mangle that into Hello, world!. It
> did not look like I had any control over the tags at all.
>
>
And
even if not, what you will do is ... code your own webframework.
>>>
>>> That is why I am looking for a class that handles the backend stuff,
>>> but lets _me_ handle the HTML.
>>
>> For God's sake : *why on earth do you believe that using a templating
>> system will make you loose _any_ control on the HTML code ???*
>>
>
> Because that's what I've seen of them.
>
>
And at least pylons/TG2 lets you return whatever you want instead, as a
string. Not via "print" though - which is simply only for CGI, and no
other
means (e.g. mod_wsgi) of python-web-programming.
>>>
>>> I am trying to follow you here. What is "return whatever you want"?
>>> Return HTML to stdout to be sent to the browser?
>>
>> please leave stdout out of here - it's HTTP, so what we have are HTTP
>> requests and HTTP responses - not stdin nor stdout. Using these streams
>> as a mean of communication between the web server and the applicative
>> code is just plain old low-level GCI stuff. Your application code
>> shouldn't have any knowledge of this implementation detail - it should
>> receive (a suitable representation of) an HTTP request, and generate (a
>> suitable representation of) an HTTP response.
>
> Can you provide a code example of printing the variable "torvalds" in
> the GET request?
> http://example.com/page.py?torvalds=tux
> Should return this:
> tux
>
> And something similar for a POST request?
>
> I hate to ask for code, but this conversation would be easier that way.
> Thanks.
I've already given you that for TG:
class RootController(BaseController):
@expose()
def page(self, torwalds=None):
return "%s" % (torwalds if torwalds
else ""
Of course nobody would do it that way. Instead, you'd define a template
${torvalds}
And then your code becomes:
@expose("your.template.path")
def page(self, torwalds=None):
return dict(torwalds=torwalds)
Separation of code from HTML-code. Which most people agree is a good thing.
And no worries about having to make torwalds a string.
Now let's say you want torwalds to be not-empty, and a number:
@expose("your.template.path")
@validate(dict(torwalds=Int(not_empty=True)),
error_handler=some_error_handler_controller_action)
def page(self, torwalds=None):
return dict(torwalds=torwalds)
Diez
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> I insist on handling the HTML myself. As for converting the request >> variables into Python variables, if a class/framework makes that >> easier then I would gladly use it. My question was serious. How can I >> do those things? > > Using a framework? In Pylons/TG2, my code looks like this: > > def some_form_action(self, name, email, password): > > > > so HTTP-variables get parsed, validated, converted, and *then* passed to my > actual code. > I know how to call a function, but how do I use the variables? How do I respond (I am avoiding the term output) with the value of the GET variable "torvalds", for example? And similarly for a POST variable? >>> But then do you deal with headers correctly? >> >> Yes, so far as I know. This is actually simpler than the HTML, just be >> careful not to output two newline characters in sequence (thereby >> ending the header). > > Oh my god. If two newlines are all you think that there is to properly > writing HTTP-headers, no wonder *you* wonder what to use a webframework > for. > I was being brief. I don't pretend to know the whole protocol by heart, but I get by that far. :) >>> Do you respect character >>> encodings? >> >> Yes! UTF-8 from database to scripting language to HTTP request. > > And if the browser sends them in different encoding? Frameworks make you > deals with unicode only. Decoding/Encoding, setting the necessary > That's what I wrote below: >>> Form-encodings? >> >> Yes, UTF-8 the in the other direction. However, as form data can be >> spoofed, I would like a function that checks it. Does Python have such >> a function? What class does? > > Frameworks do check for HTTP-headers that contain the encoding, and thus > convert incoming data to unicode. > That is useful, thanks. >> >> >>> Is your generated HTML valid? >> >> Naturally, even though this is not a Python issue. > > "Naturally" with print-statements? > "Carefully" with print statements. >>> Do you parse >>> content negotiation headers? >>> >> >> No. I hate sites that do that. If the page is available in another >> language, their is a link in the corner. > > Again, you don't seem to really know much about webapps. Content-negotiation > isn't about languages, it's about content-types. Such as serving HTML or > JSON from the same URL. But let me guess, you hate applications that do > that? > I used to serve WML-specific "decks" to known mobile UA's, but I've stopped. > I don't have the *slightest* doubt about that. In fact, I'm pretty sure you > can make pretty bad webapplications in pretty much everything you touch. > Good luck with that. > I am a mechanical engineering student, not CS. I am doing this for my own use and enjoyment. I don't need luck with that, I don't have any customers. If someone were to come along asking me to code for them, I would send them to you! -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> I insist on handling the HTML myself. > > I just don't get how embedding HTML in applicative code - a well-known > antipattern FWIW - relates to "handling the HTML yourself". Can you *please* > explain how a templating system prevents you from "handling the HTML" > yourself. > >From the little that I played with Django, it looked like I would (for the sake of argument I am just using the term output, but the intention should be clear) output "Hello, world!" and Django would mangle that into Hello, world!. It did not look like I had any control over the tags at all. >>> And >>> even if not, what you will do is ... code your own webframework. >> >> That is why I am looking for a class that handles the backend stuff, >> but lets _me_ handle the HTML. > > For God's sake : *why on earth do you believe that using a templating system > will make you loose _any_ control on the HTML code ???* > Because that's what I've seen of them. >>> And at least pylons/TG2 lets you return whatever you want instead, as a >>> string. Not via "print" though - which is simply only for CGI, and no >>> other >>> means (e.g. mod_wsgi) of python-web-programming. >>> >> >> I am trying to follow you here. What is "return whatever you want"? >> Return HTML to stdout to be sent to the browser? > > please leave stdout out of here - it's HTTP, so what we have are HTTP > requests and HTTP responses - not stdin nor stdout. Using these streams as a > mean of communication between the web server and the applicative code is > just plain old low-level GCI stuff. Your application code shouldn't have any > knowledge of this implementation detail - it should receive (a suitable > representation of) an HTTP request, and generate (a suitable representation > of) an HTTP response. Can you provide a code example of printing the variable "torvalds" in the GET request? http://example.com/page.py?torvalds=tux Should return this: tux And something similar for a POST request? I hate to ask for code, but this conversation would be easier that way. Thanks. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
While I know that to be true in the general sense, from what I've looked at Django and other frameworks it seems that the web frameworks push the coder to use templates, not letting him near the HTML. >>> >>> Well... there must be a reason, for sure... No ? >>> >> >> ֹYes, but I don't like it. > > Why so ? What's your problem with templating systems, exactly ? > They don't give me the control over the HTML that I would like. I'm probably stuck in 1998 when they didn't exist, but I like to write the HTML myself. Django and the other frameworks seem to force the user to use templates. >>> >>> Not at all. Nothing prevents you from shooting yourself in the foot and >>> generating HTML "by hand" in your view functions (or request handler >>> methods >>> etc, depending on the framework). >> >> How is this done in Django, then? > > def index(request): > unmaintanable_html = """ > > > Index > > > Embedded HTML is a PITA > but some like pains... > > > """ > return HttpResponse(unmaintanable_html) > And if I need to add a variable or three in there? Static HTML I can do without Python. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> I have no idea what reverse url generation is. > > It's a mean to avoid hard-coding urls in your application code - the url is > fully generated by the 'url mapping' system. > I don't need that, see below. >> I assume that the user >> will call http://example.com/path/to/script.py?var1=hello&var2=world >> >> script.py would live in /home/user/site-name/public_html/path/to/ >> > > Now try to use apache url rewrites to use "nice urls" (like, > "///" instead of > "path/to/script-serving-articles.wtf?id=42"). Good luck. > Actually, currently in the example url http://example.com/path/to/script.py?var1=hello&var2=world the script is /home/user/site-name/public_html/path/ (with no filename extension) and the script.py is actually page.html which is an arbitrary, descriptive string and not part of the script filename. See the pages on http://dotancohen.com for examples. >> I would prefer to output everything from to with print >> statements. I don't want some framework wrapping my output in tags, I >> want the tags to be part of the output. > > You're confusing the framework and the templating system. The framework by > itself won't wrap your "output" in anything, nor even forces you to "output" > HTML. This is what I was hoping to hear. I don't remember the details, but I had a hard time outputting the HTML myself in Django. I will play with it again soon if I get the chance. > Also FWIW, it's HTTP, so we don't have "outputs", we have HTTP > responses. > Clearly. I was referring to stdout being send to the browser as the http response. I think I mentioned that, but I apologize for being unclear. >> Yes, why not? > > Let's say I want to reuse your app with a different presentation. Do I have > to fork the whole application and send my HTML coder dig into your > applicative code just to rewrite the HTML parts ? > No, just replace two files. In that manner it is templating. Again, an example from the current PHP code: >> Like I said before, I don't want to have to maintain the functions >> that turn the HTTP environment into Python variables, > > "HTTP environment" ??? > > Oh, you mean HTTP requests etc... Well, the cgi module already provides some > basic support here. > Does it? I will look into that. I assume that basic support means making the cookie, GET and POST variables easily accessible. I google, but cannot find any exapmles of this online. >> or the code that >> manages database connections. Functions that escape data to be sent to >> MySQL (to prevent sql injection) would be nice. > > Any DB-API compliant module already provide this - granted you use it > correctly. > I know. This is will use correctly, I promise! >> Other than that, it's >> all on me. > > Your choice... > Hey! Since when am I not under attack? :) -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen wrote: >>> I should probably expand on this: >>> >>> How can I get an array with all the GET variables in Python? >>> How can I get an array with all the POST variables in Python? >>> How can I get an array with all the COOKIE variables in Python? >>> How can I get the request URI path (everything after >>> http://[www.?]example.com/)? >>> >>> That's all I want: no templates and nothing between me and the HTML. >>> The HTTP headers I can output to stdout myself as well. >> >> Again: if you insist on doing everything yourself - then of course any >> library or framework isn't for you. >> > > I insist on handling the HTML myself. As for converting the request > variables into Python variables, if a class/framework makes that > easier then I would gladly use it. My question was serious. How can I > do those things? Using a framework? In Pylons/TG2, my code looks like this: def some_form_action(self, name, email, password): so HTTP-variables get parsed, validated, converted, and *then* passed to my actual code. >> But then do you deal with headers correctly? > > Yes, so far as I know. This is actually simpler than the HTML, just be > careful not to output two newline characters in sequence (thereby > ending the header). Oh my god. If two newlines are all you think that there is to properly writing HTTP-headers, no wonder *you* wonder what to use a webframework for. http://www.w3.org/Protocols/rfc2616/rfc2616-sec5.html#sec5.3 http://www.w3.org/Protocols/rfc2616/rfc2616-sec6.html#sec6.2 That's just an *overwiew* of what different kinds of headers there are, all of which might require specific rules to parse or generate. > >> Do you respect character >> encodings? > > Yes! UTF-8 from database to scripting language to HTTP request. And if the browser sends them in different encoding? Frameworks make you deals with unicode only. Decoding/Encoding, setting the necessary >> Form-encodings? > > Yes, UTF-8 the in the other direction. However, as form data can be > spoofed, I would like a function that checks it. Does Python have such > a function? What class does? Frameworks do check for HTTP-headers that contain the encoding, and thus convert incoming data to unicode. > > >> Is your generated HTML valid? > > Naturally, even though this is not a Python issue. "Naturally" with print-statements? I use genshi templating, which is XML-based, so I can't even write invalid HTML for a great deal of cases. > > >> Are >> timestamp-formats generated according to RFCs for your cookies? > > Yes, this is not a problem. Is there some gothcha that I am unaware of? The thing is that using a framework, I just pass a datetime-object. Others have done it right for me translating that to the needed format. > > >> Do you parse >> content negotiation headers? >> > > No. I hate sites that do that. If the page is available in another > language, their is a link in the corner. Again, you don't seem to really know much about webapps. Content-negotiation isn't about languages, it's about content-types. Such as serving HTML or JSON from the same URL. But let me guess, you hate applications that do that? > > >> I think you underestimate the task it is to make a webapplication good. > > Probably, but that would not depend on the scripting language. I make > bad webapplications in PHP too! I don't have the *slightest* doubt about that. In fact, I'm pretty sure you can make pretty bad webapplications in pretty much everything you touch. Good luck with that. Diez -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : I should probably expand on this: How can I get an array with all the GET variables in Python? How can I get an array with all the POST variables in Python? How can I get an array with all the COOKIE variables in Python? How can I get the request URI path (everything after http://[www.?]example.com/)? That's all I want: no templates and nothing between me and the HTML. The HTTP headers I can output to stdout myself as well. Again: if you insist on doing everything yourself - then of course any library or framework isn't for you. I insist on handling the HTML myself. I just don't get how embedding HTML in applicative code - a well-known antipattern FWIW - relates to "handling the HTML yourself". Can you *please* explain how a templating system prevents you from "handling the HTML" yourself. I think you underestimate the task it is to make a webapplication good. Probably, but that would not depend on the scripting language. I make bad webapplications in PHP too! Lol ! And even if not, what you will do is ... code your own webframework. That is why I am looking for a class that handles the backend stuff, but lets _me_ handle the HTML. For God's sake : *why on earth do you believe that using a templating system will make you loose _any_ control on the HTML code ???* And at least pylons/TG2 lets you return whatever you want instead, as a string. Not via "print" though - which is simply only for CGI, and no other means (e.g. mod_wsgi) of python-web-programming. I am trying to follow you here. What is "return whatever you want"? Return HTML to stdout to be sent to the browser? please leave stdout out of here - it's HTTP, so what we have are HTTP requests and HTTP responses - not stdin nor stdout. Using these streams as a mean of communication between the web server and the applicative code is just plain old low-level GCI stuff. Your application code shouldn't have any knowledge of this implementation detail - it should receive (a suitable representation of) an HTTP request, and generate (a suitable representation of) an HTTP response. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : While I know that to be true in the general sense, from what I've looked at Django and other frameworks it seems that the web frameworks push the coder to use templates, not letting him near the HTML. Well... there must be a reason, for sure... No ? ֹYes, but I don't like it. Why so ? What's your problem with templating systems, exactly ? Django and the other frameworks seem to force the user to use templates. Not at all. Nothing prevents you from shooting yourself in the foot and generating HTML "by hand" in your view functions (or request handler methods etc, depending on the framework). How is this done in Django, then? def index(request): unmaintanable_html = """ Index Embedded HTML is a PITA but some like pains... """ return HttpResponse(unmaintanable_html) This is just plain stupid wrt/ readability, maintainability, reuse and whatnot, but hey, if you want to waste your time, please do. As far as i'm concerned, I'm very happy to let the HTML coder write the HTML part and the application programmer write the applicative part - even when I end up wearing both caps. I've done stupider things. Your choice, once again... -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : 2009/10/28 Bruno Desthuilliers : Dotan Cohen a écrit : declarative mapping of urls to code Apache does this, unless I am misunderstanding you. Using url rewriting ? Yes, fine. Then tell me how you implement "reverse" url generation (like Django or Routes do). I have no idea what reverse url generation is. It's a mean to avoid hard-coding urls in your application code - the url is fully generated by the 'url mapping' system. I assume that the user will call http://example.com/path/to/script.py?var1=hello&var2=world script.py would live in /home/user/site-name/public_html/path/to/ Now try to use apache url rewrites to use "nice urls" (like, "///" instead of "path/to/script-serving-articles.wtf?id=42"). Good luck. of code to templates Those who code in HTML don't need this. ??? I would prefer to output everything from to with print statements. I don't want some framework wrapping my output in tags, I want the tags to be part of the output. You're confusing the framework and the templating system. The framework by itself won't wrap your "output" in anything, nor even forces you to "output" HTML. Also FWIW, it's HTTP, so we don't have "outputs", we have HTTP responses. In any case it's not hard to call a function in a class that writes the HTML before the content, then write the content, then call another function that writes the HTML after the content. Hmmm, yummy ! And SO maintainable... Yes, why not? Let's say I want to reuse your app with a different presentation. Do I have to fork the whole application and send my HTML coder dig into your applicative code just to rewrite the HTML parts ? This is how my sites are run, though in PHP instead of Python. No prepackaged templates. PHP *is* a template language. (snip) I would really like to know what else. So far, I am not convinced that a framework offers anything that is not already easily accomplished in Python. Given that we're talking about Python frameworks, it seems obvious that what they do can be accomplished in Python. Now the question is how much you like to write the same boring and error-prone boilerplate code project after project... Like I said before, I don't want to have to maintain the functions that turn the HTTP environment into Python variables, "HTTP environment" ??? Oh, you mean HTTP requests etc... Well, the cgi module already provides some basic support here. or the code that manages database connections. Functions that escape data to be sent to MySQL (to prevent sql injection) would be nice. Any DB-API compliant module already provide this - granted you use it correctly. Other than that, it's all on me. Your choice... -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> How can I get an array with all the GET variables in Python?
>> How can I get an array with all the POST variables in Python?
>> How can I get an array with all the COOKIE variables in Python?
>> How can I get the request URI path (everything after
>> http://[www.?]example.com/)?
>>
>> That's all I want: no templates and nothing between me and the HTML.
>> The HTTP headers I can output to stdout myself as well.
>
> $python
>>>help(cgi)
>>>help(cgitb)
>>>help(Cookie)
>>>help(urlparse)
>>>help(os.environ)
>
> GET variables:
> -
> import os,cgi
> if os.environ.has_key('QUERY_STRING'):
> getv = cgi.parse_qs(os.environ['QUERY_STRING'])
> print getv.keys()
>
> Other examples is very simple too...
>
Thanks, Mikhail. I now have a bit to google on.
--
Dotan Cohen
http://what-is-what.com
http://gibberish.co.il
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> While I know that to be true in the general sense, from what I've >> looked at Django and other frameworks it seems that the web frameworks >> push the coder to use templates, not letting him near the HTML. > > Well... there must be a reason, for sure... No ? > ֹYes, but I don't like it. >> Django and the other frameworks seem to >> force the user to use templates. > > Not at all. Nothing prevents you from shooting yourself in the foot and > generating HTML "by hand" in your view functions (or request handler methods > etc, depending on the framework). How is this done in Django, then? > This is just plain stupid wrt/ > readability, maintainability, reuse and whatnot, but hey, if you want to > waste your time, please do. As far as i'm concerned, I'm very happy to let > the HTML coder write the HTML part and the application programmer write the > applicative part - even when I end up wearing both caps. > I've done stupider things. >> I just want the functions, and to >> print the HTML as stdout to the browser making the request. I had to >> settle on PHP to do this, > > Why so ? Almost any programming language can do CGI. And I settled on PHP because it does what I need. However, I would prefer to move it to Python for the benefit of using the same language in the different places that I code for (such as personal applications for the desktop, pyqt). -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
2009/10/28 Bruno Desthuilliers : > Dotan Cohen a écrit : >>> >>> declarative mapping of urls to code >> >> Apache does this, unless I am misunderstanding you. > > Using url rewriting ? Yes, fine. Then tell me how you implement "reverse" > url generation (like Django or Routes do). I have no idea what reverse url generation is. I assume that the user will call http://example.com/path/to/script.py?var1=hello&var2=world script.py would live in /home/user/site-name/public_html/path/to/ >>> of code to templates >> >> Those who code in HTML don't need this. > > ??? > I would prefer to output everything from to with print statements. I don't want some framework wrapping my output in tags, I want the tags to be part of the output. >> In any case it's not hard to >> call a function in a class that writes the HTML before the content, >> then write the content, then call another function that writes the >> HTML after the content. > > Hmmm, yummy ! And SO maintainable... > Yes, why not? >> This is how my sites are run, though in PHP >> instead of Python. No prepackaged templates. > > PHP *is* a template language. > This is, in general what I'm doing: >> I would really like to know what else. So far, I am not convinced that >> a framework offers anything that is not already easily accomplished in >> Python. > > Given that we're talking about Python frameworks, it seems obvious that what > they do can be accomplished in Python. Now the question is how much you like > to write the same boring and error-prone boilerplate code project after > project... > Like I said before, I don't want to have to maintain the functions that turn the HTTP environment into Python variables, or the code that manages database connections. Functions that escape data to be sent to MySQL (to prevent sql injection) would be nice. Other than that, it's all on me. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
>> I should probably expand on this: >> >> How can I get an array with all the GET variables in Python? >> How can I get an array with all the POST variables in Python? >> How can I get an array with all the COOKIE variables in Python? >> How can I get the request URI path (everything after >> http://[www.?]example.com/)? >> >> That's all I want: no templates and nothing between me and the HTML. >> The HTTP headers I can output to stdout myself as well. > > Again: if you insist on doing everything yourself - then of course any > library or framework isn't for you. > I insist on handling the HTML myself. As for converting the request variables into Python variables, if a class/framework makes that easier then I would gladly use it. My question was serious. How can I do those things? > But then do you deal with headers correctly? Yes, so far as I know. This is actually simpler than the HTML, just be careful not to output two newline characters in sequence (thereby ending the header). > Do you respect character > encodings? Yes! UTF-8 from database to scripting language to HTTP request. > Form-encodings? Yes, UTF-8 the in the other direction. However, as form data can be spoofed, I would like a function that checks it. Does Python have such a function? What class does? > Is your generated HTML valid? Naturally, even though this is not a Python issue. > Are > timestamp-formats generated according to RFCs for your cookies? Yes, this is not a problem. Is there some gothcha that I am unaware of? > Do you parse > content negotiation headers? > No. I hate sites that do that. If the page is available in another language, their is a link in the corner. > I think you underestimate the task it is to make a webapplication good. Probably, but that would not depend on the scripting language. I make bad webapplications in PHP too! > And > even if not, what you will do is ... code your own webframework. That is why I am looking for a class that handles the backend stuff, but lets _me_ handle the HTML. > Because > there is a lot of boilerplate otherwis. If that's a learning-experience your > after, fine. > > Besides, yes, you can get all these things nonetheless. You just don't need > them most of the time. > > And at least pylons/TG2 lets you return whatever you want instead, as a > string. Not via "print" though - which is simply only for CGI, and no other > means (e.g. mod_wsgi) of python-web-programming. > I am trying to follow you here. What is "return whatever you want"? Return HTML to stdout to be sent to the browser? -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Hello, Dotan
On Wed, 28 Oct 2009 10:26:22 +0200
Dotan Cohen wrote:
> I should probably expand on this:
>
> How can I get an array with all the GET variables in Python?
> How can I get an array with all the POST variables in Python?
> How can I get an array with all the COOKIE variables in Python?
> How can I get the request URI path (everything after
> http://[www.?]example.com/)?
>
> That's all I want: no templates and nothing between me and the HTML.
> The HTTP headers I can output to stdout myself as well.
$python
>>help(cgi)
>>help(cgitb)
>>help(Cookie)
>>help(urlparse)
>>help(os.environ)
GET variables:
-
import os,cgi
if os.environ.has_key('QUERY_STRING'):
getv = cgi.parse_qs(os.environ['QUERY_STRING'])
print getv.keys()
Other examples is very simple too...
--
Mikhail M.Smagin
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
John Nagle a écrit : Alan Harris-Reid wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 Until MySQLdb gets ported to something later than Python 2.5, support for a "data-based web site" probably has to be in Python 2.5 or earlier. hem br...@bruno:~$ python Python 2.6.2 (release26-maint, Apr 19 2009, 01:56:41) [GCC 4.3.3] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import MySQLdb >>> MySQLdb '/var/lib/python-support/python2.6/MySQLdb/__init__.pyc'> You said ??? (snip) Frameworks are more useful if you're doing yet another "Web 2.0" web site. I beg to disagree. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : A webframework is *written* in python. Your whole line of argumentation boils down to "I can write things myself in python, why use libraries/frameworks". Yes. You can also delete your standard-lib, and code everything in there yourself - with the same argument. Using a framework is about proven solutions for common problems, letting you focus on working on your actual application code. While I know that to be true in the general sense, from what I've looked at Django and other frameworks it seems that the web frameworks push the coder to use templates, not letting him near the HTML. Well... there must be a reason, for sure... No ? Django and the other frameworks seem to force the user to use templates. Not at all. Nothing prevents you from shooting yourself in the foot and generating HTML "by hand" in your view functions (or request handler methods etc, depending on the framework). This is just plain stupid wrt/ readability, maintainability, reuse and whatnot, but hey, if you want to waste your time, please do. As far as i'm concerned, I'm very happy to let the HTML coder write the HTML part and the application programmer write the applicative part - even when I end up wearing both caps. I just want the functions, and to print the HTML as stdout to the browser making the request. I had to settle on PHP to do this, Why so ? Almost any programming language can do CGI. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : Well, yes- but it's also DRY, and while DRTW (like the acronym, btw) helps to prevent your code from being unreadable to someone else, DRY helps to ensure that when you have to change something you don't have to worry about changing it in 37 and a half other places at the same time. Especially given how notoriously difficult it is to do automated testing for web development, that's essential. That's what classes are for, no? Reuse code. Nope. "classes" are an artifact of one possible approach of OOP, period. I fail to see how using a framework would reduce the need to use classes, Why should it ? -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen a écrit : declarative mapping of urls to code Apache does this, unless I am misunderstanding you. Using url rewriting ? Yes, fine. Then tell me how you implement "reverse" url generation (like Django or Routes do). of code to templates Those who code in HTML don't need this. ??? In any case it's not hard to call a function in a class that writes the HTML before the content, then write the content, then call another function that writes the HTML after the content. Hmmm, yummy ! And SO maintainable... This is how my sites are run, though in PHP instead of Python. No prepackaged templates. PHP *is* a template language. (snip) I would really like to know what else. So far, I am not convinced that a framework offers anything that is not already easily accomplished in Python. Given that we're talking about Python frameworks, it seems obvious that what they do can be accomplished in Python. Now the question is how much you like to write the same boring and error-prone boilerplate code project after project... -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
John Nagle schrieb: Alan Harris-Reid wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 Until MySQLdb gets ported to something later than Python 2.5, support for a "data-based web site" probably has to be in Python 2.5 or earlier. The usual FUD from you. http://pypi.python.org/pypi/MySQL-python/1.2.3c1 And of course there are other alternatives to databases, which happily run under 2.6 as well. Diez -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen schrieb: While I know that to be true in the general sense, from what I've looked at Django and other frameworks it seems that the web frameworks push the coder to use templates, not letting him near the HTML. For instance, I was looking for a class / framework that provided a proven method of decoding cookies (setting them is no problem), decoding POST and GET variables, escaping variables for safe entry into MySQL, and other things. Django and the other frameworks seem to force the user to use templates. I just want the functions, and to print the HTML as stdout to the browser making the request. I had to settle on PHP to do this, which admittedly is what PHP was invented to do. However, for obvious reasons, I would have prefered to code in Python. In fact, I still would. I should probably expand on this: How can I get an array with all the GET variables in Python? How can I get an array with all the POST variables in Python? How can I get an array with all the COOKIE variables in Python? How can I get the request URI path (everything after http://[www.?]example.com/)? That's all I want: no templates and nothing between me and the HTML. The HTTP headers I can output to stdout myself as well. Again: if you insist on doing everything yourself - then of course any library or framework isn't for you. But then do you deal with headers correctly? Do you respect character encodings? Form-encodings? Is your generated HTML valid? Are timestamp-formats generated according to RFCs for your cookies? Do you parse content negotiation headers? I think you underestimate the task it is to make a webapplication good. And even if not, what you will do is ... code your own webframework. Because there is a lot of boilerplate otherwis. If that's a learning-experience your after, fine. Besides, yes, you can get all these things nonetheless. You just don't need them most of the time. And at least pylons/TG2 lets you return whatever you want instead, as a string. Not via "print" though - which is simply only for CGI, and no other means (e.g. mod_wsgi) of python-web-programming. Diez -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> Even 2.6 is too bleeding-edge? Uh oh, Fedora 12 is shipping it. > Fedora has traditionally been known as a bleeding edge distro. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> While I know that to be true in the general sense, from what I've > looked at Django and other frameworks it seems that the web frameworks > push the coder to use templates, not letting him near the HTML. > > For instance, I was looking for a class / framework that provided a > proven method of decoding cookies (setting them is no problem), > decoding POST and GET variables, escaping variables for safe entry > into MySQL, and other things. Django and the other frameworks seem to > force the user to use templates. I just want the functions, and to > print the HTML as stdout to the browser making the request. I had to > settle on PHP to do this, which admittedly is what PHP was invented to > do. However, for obvious reasons, I would have prefered to code in > Python. In fact, I still would. > I should probably expand on this: How can I get an array with all the GET variables in Python? How can I get an array with all the POST variables in Python? How can I get an array with all the COOKIE variables in Python? How can I get the request URI path (everything after http://[www.?]example.com/)? That's all I want: no templates and nothing between me and the HTML. The HTTP headers I can output to stdout myself as well. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
John Nagle writes: > Until MySQLdb gets ported to something later than Python 2.5, support > for a "data-based web site" probably has to be in Python 2.5 or earlier. Even 2.6 is too bleeding-edge? Uh oh, Fedora 12 is shipping it. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> A webframework is *written* in python. Your whole line of argumentation > boils down to "I can write things myself in python, why use > libraries/frameworks". Yes. You can also delete your standard-lib, and code > everything in there yourself - with the same argument. > > Using a framework is about proven solutions for common problems, letting you > focus on working on your actual application code. > While I know that to be true in the general sense, from what I've looked at Django and other frameworks it seems that the web frameworks push the coder to use templates, not letting him near the HTML. For instance, I was looking for a class / framework that provided a proven method of decoding cookies (setting them is no problem), decoding POST and GET variables, escaping variables for safe entry into MySQL, and other things. Django and the other frameworks seem to force the user to use templates. I just want the functions, and to print the HTML as stdout to the browser making the request. I had to settle on PHP to do this, which admittedly is what PHP was invented to do. However, for obvious reasons, I would have prefered to code in Python. In fact, I still would. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Alan Harris-Reid wrote:
I am very much new to Python, and one of my first projects is a simple
data-based website. I am starting with Python 3.1
Until MySQLdb gets ported to something later than Python 2.5, support
for a "data-based web site" probably has to be in Python 2.5 or earlier.
The C module situation for Python 3.x still isn't very good.
Realistically, the production version of Python is 2.5. This process
is taking long enough that I'm worried that Python 3.x could do for
Python what Perl 6 did for Perl - provide an upgrade path that nobody
takes.
HTMLTemplate ("http://py-templates.sourceforge.net/htmltemplate/index.html";)
is a minimal templating system for fill-in-the-blanks template work in Python.
Actually, if you have HTMLTemplate, FCGI, and MySQLdb, you have enough to
do a back-end database.
There are some advantages to libraries (you call them) over "frameworks"
(they call you) if you're doing something unusual. Frameworks are more
useful if you're doing yet another "Web 2.0" web site.
John Nagle
--
http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen schrieb: Well, yes- but it's also DRY, and while DRTW (like the acronym, btw) helps to prevent your code from being unreadable to someone else, DRY helps to ensure that when you have to change something you don't have to worry about changing it in 37 and a half other places at the same time. Especially given how notoriously difficult it is to do automated testing for web development, that's essential. That's what classes are for, no? Reuse code. I fail to see how using a framework would reduce the need to use classes, provided that the need exists. A webframework is *written* in python. Your whole line of argumentation boils down to "I can write things myself in python, why use libraries/frameworks". Yes. You can also delete your standard-lib, and code everything in there yourself - with the same argument. Using a framework is about proven solutions for common problems, letting you focus on working on your actual application code. Diez -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Tue, Oct 27, 2009 at 4:52 PM, Dotan Cohen wrote: >> Well, yes- but it's also DRY, and while DRTW (like the acronym, btw) >> helps to prevent your code from being unreadable to someone else, >> DRY helps to ensure that when you have to change something you >> don't have to worry about changing it in 37 and a half other places >> at the same time. Especially given how notoriously difficult it is to >> do automated testing for web development, that's essential. >> > > That's what classes are for, no? Reuse code. I fail to see how using a > framework would reduce the need to use classes, provided that the need > exists. ...frameworks use classes. They just don't make you write all of them. Geremy Condra -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> Well, yes- but it's also DRY, and while DRTW (like the acronym, btw) > helps to prevent your code from being unreadable to someone else, > DRY helps to ensure that when you have to change something you > don't have to worry about changing it in 37 and a half other places > at the same time. Especially given how notoriously difficult it is to > do automated testing for web development, that's essential. > That's what classes are for, no? Reuse code. I fail to see how using a framework would reduce the need to use classes, provided that the need exists. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Tue, Oct 27, 2009 at 3:11 PM, Dotan Cohen wrote: >> Using a framework helps to ensure that your code is easy to maintain. > > I see, that is a good point. With someone else (the framework > maintainers) worrying about maintaining function such as HTTP request > parsers, a lot of work won't have to be [re]done. > > >> DRY isn't about saving time now, its about saving time six months >> from now. >> > > I suppose in this case it's DRTW (don't reinvent the wheel) but the > same principle applies. Well, yes- but it's also DRY, and while DRTW (like the acronym, btw) helps to prevent your code from being unreadable to someone else, DRY helps to ensure that when you have to change something you don't have to worry about changing it in 37 and a half other places at the same time. Especially given how notoriously difficult it is to do automated testing for web development, that's essential. Geremy Condra -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Oct 27, 10:26 am, "Diez B. Roggisch" wrote: ... > Yes, in the end of the day, it's all python. For me, in the end of the day, it's all java or PHP. But I'm working on that. For my purposes the "frameworks" don't really help much. That's why I built WHIFF :). http://aaron.oirt.rutgers.edu/myapp/docs/W1100_2200.TreeView -- Aaron Watters === If all you got is lemons, make lemonade. -- anon. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> Using a framework helps to ensure that your code is easy to maintain. I see, that is a good point. With someone else (the framework maintainers) worrying about maintaining function such as HTTP request parsers, a lot of work won't have to be [re]done. > DRY isn't about saving time now, its about saving time six months > from now. > I suppose in this case it's DRTW (don't reinvent the wheel) but the same principle applies. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Tue, Oct 27, 2009 at 2:36 PM, Dotan Cohen wrote: >> declarative mapping of urls to code > > Apache does this, unless I am misunderstanding you. > > >> of code to templates > > Those who code in HTML don't need this. In any case it's not hard to > call a function in a class that writes the HTML before the content, > then write the content, then call another function that writes the > HTML after the content. This is how my sites are run, though in PHP > instead of Python. No prepackaged templates. > >> abstracting away >> the details of GET and POST > > GET is easy, just parse the HTTP request. I don't know how much of a > problem POST would be. > > >> validating and decoding parameters, especially >> if these become larger repetitive structures like several addresses of a >> user > > This falls under database, cookies, or HTTP request parsing. Or am I > misunderstanding something again? > > >> re-rendering invalid form-data > > Just add it into the HTML. > > >> working with HTML or JSON as output, > > Same as writing to stdout, just output the HTTP headers first. > > >> managing transactions, providing a error-reporting-infrastructure >> > > This does not differ from regular (non-web) Python coding. > > >> The list continues. >> > > I would really like to know what else. So far, I am not convinced that > a framework offers anything that is not already easily accomplished in > Python. Using a framework helps to ensure that your code is easy to maintain. DRY isn't about saving time now, its about saving time six months from now. Geremy Condra -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> declarative mapping of urls to code Apache does this, unless I am misunderstanding you. > of code to templates Those who code in HTML don't need this. In any case it's not hard to call a function in a class that writes the HTML before the content, then write the content, then call another function that writes the HTML after the content. This is how my sites are run, though in PHP instead of Python. No prepackaged templates. > abstracting away > the details of GET and POST GET is easy, just parse the HTTP request. I don't know how much of a problem POST would be. > validating and decoding parameters, especially > if these become larger repetitive structures like several addresses of a > user This falls under database, cookies, or HTTP request parsing. Or am I misunderstanding something again? > re-rendering invalid form-data Just add it into the HTML. > working with HTML or JSON as output, Same as writing to stdout, just output the HTTP headers first. > managing transactions, providing a error-reporting-infrastructure > This does not differ from regular (non-web) Python coding. > The list continues. > I would really like to know what else. So far, I am not convinced that a framework offers anything that is not already easily accomplished in Python. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> I am very much new to Python, and one of my first projects is a simple > data-based website. I am starting with Python 3.1 (I can hear many of > you shouting "don't - start with 2.6"), but as far as I can see, none of > the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) > are Python3 compatible yet. That's not entirely true, see http://wiki.python.org/moin/PortingDjangoTo3k Regards, Martin -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Dotan Cohen wrote: > Why the push to use a framework, and why the resistance from the OP? > > Does the OP need to work with cookies or other http-specific features? > In fact, other than cookies, what http-specific features exist? declarative mapping of urls to code, of code to templates, abstracting away the details of GET and POST, validating and decoding parameters, especially if these become larger repetitive structures like several addresses of a user, re-rendering invalid form-data, working with HTML or JSON as output, managing transactions, providing a error-reporting-infrastructure The list continues. > Does > Python have a built-in framework for making available GET and POST > variables? Database queries should be no different than in other > Python apps. Yes, in the end of the day, it's all python. Diez -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Why the push to use a framework, and why the resistance from the OP? Does the OP need to work with cookies or other http-specific features? In fact, other than cookies, what http-specific features exist? Does Python have a built-in framework for making available GET and POST variables? Database queries should be no different than in other Python apps. -- Dotan Cohen http://what-is-what.com http://gibberish.co.il -- http://mail.python.org/mailman/listinfo/python-list
RE: Web development with Python 3.1
I would agree with Alan. Most of the libraries you should use are compatible with the 2.x series.. I still use versions 2.5 and 2.6 for all development. From: python-list-bounces+billy.earney=gmail@python.org [mailto:python-list-bounces+billy.earney=gmail@python.org] On Behalf Of Alan Harris-Reid Sent: Tuesday, October 27, 2009 6:08 AM To: Python List Subject: Re: Web development with Python 3.1 Aaron Watters wrote: On Oct 25, 7:52 pm, Alan Harris-Reid <mailto:a...@baselinedata.co.uk> wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 (I can hear many of you shouting "don't - start with 2.6"), but as far as I can see, none of the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are Python3 compatible yet. So, what can I use to start my web programming experience using 3.1? Any help would be appreciated. Alan Don't. use python 2.6 with WHIFF :) http://aaron.oirt.rutgers.edu/myapp/GenBankTree/index http://whiff.sourceforge.net -- Aaron Watters Thanks for the advice Alan -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Aaron Watters wrote: On Oct 25, 7:52 pm, Alan Harris-Reid wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 (I can hear many of you shouting "don't - start with 2.6"), but as far as I can see, none of the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are Python3 compatible yet. So, what can I use to start my web programming experience using 3.1? Any help would be appreciated. Alan Don't. use python 2.6 with WHIFF :) http://aaron.oirt.rutgers.edu/myapp/GenBankTree/index http://whiff.sourceforge.net -- Aaron Watters Thanks for the advice Alan -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On Oct 25, 7:52 pm, Alan Harris-Reid wrote: > I am very much new to Python, and one of my first projects is a simple > data-based website. I am starting with Python 3.1 (I can hear many of > you shouting "don't - start with 2.6"), but as far as I can see, none of > the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) > are Python3 compatible yet. > > So, what can I use to start my web programming experience using 3.1? > > Any help would be appreciated. > > Alan Don't. use python 2.6 with WHIFF :) http://aaron.oirt.rutgers.edu/myapp/GenBankTree/index http://whiff.sourceforge.net -- Aaron Watters === It gotta be rock-roll music if you wanna dance with me if you wanna dance with me -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Hi Paul, thanks for the reply (despite the sarcasm ;-) ), >Does it occur to you that the unavailability of those frameworks is part of the REASON they say to use 2.x? Of course, but that doesn't mean that there isn't someone out there who may know of a framework that is already Python3 compatible, or is being worked-on. >Have you answered your own question? No >Anyway, for simple web programming, frameworks are not worth the hassle. Just use the cgi module. Ok - I'll look into it. >... you are the one who decided to zoom off into the 3.1 wilderness before the framework developers got there. I haven't zoomed-off anywhere yet - in fact I've hardly started. I made the decision to start with 3.1 based on the fact that it was the latest version, and hoping that there may be some framework stuff out there (because 3.0 has been out for a while now). However, it looks as though I might have to review that decision, but what I've learned so far is pretty simple stuff, so it won't be wasted. Regards, Alan Alan Harris-Reid writes: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 (I can hear many of you shouting "don't - start with 2.6"), but as far as I can see, none of the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are Python3 compatible yet. So, what can I use to start my web programming experience using 3.1? Does it occur to you that the unavailability of those frameworks is part of the REASON they say to use 2.x? Have you answered your own question? Anyway, for simple web programming, frameworks are not worth the hassle. Just use the cgi module. If you want to use a framework, well, you are the one who decided to zoom off into the 3.1 wilderness before the framework developers got there. If you're an experienced programmer in other languages and you're determined to use a framework, maybe a worthwhile Python learning project would be to help port your favorite framework to 3.1. -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Anyway, for simple web programming, frameworks are not worth the hassle. Just use the cgi module. I can vouch for what Paul says. I started in Python 3 years ago, and I did so with a web application (still working on it!). I'm using the cgi approach, and it certainly teaches you the concepts. I fail to see how starting with a framework is a good idea if you don't know how the frameworks work (or what they're actually doing). It would be a bit like doing a web page in Dreamw***er and thinking you understand HTML/CSS. B Hi Brendon, thanks for the advice. Looks like I'll have to go the cgi route myself if I want to stick with with Python3. Regards, Alan -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Exarkun - thanks for the reply > don't - start with 2.6 Thought you might say that ;-) Regards, Alan On 25 Oct, 11:52 pm, a...@baselinedata.co.uk wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 (I can hear many of you shouting "don't - start with 2.6"), but as far as I can see, none of the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are Python3 compatible yet. So, what can I use to start my web programming experience using 3.1? Any help would be appreciated. don't - start with 2.6 Alan -- http://mail.python.org/mailman/listinfo/python-list No virus found in this incoming message. Checked by AVG - www.avg.com Version: 8.5.423 / Virus Database: 270.14.31/2458 - Release Date: 10/25/09 08:10:00 -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
bottle (http://bottle.paws.de/) can run on python 3.1 after running the 2to3 tool on it. It is a very lightweight framework. CherryPy 3.2 also runs on python 3.x I don't know if there are any others. On Sun, Oct 25, 2009 at 7:52 PM, Alan Harris-Reid wrote: > > I am very much new to Python, and one of my first projects is a simple > data-based website. I am starting with Python 3.1 (I can hear many of you > shouting "don't - start with 2.6"), but as far as I can see, none of the > popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are > Python3 compatible yet. > > So, what can I use to start my web programming experience using 3.1? > > Any help would be appreciated. > > Alan > > -- > http://mail.python.org/mailman/listinfo/python-list > -- -Brandon Singer -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Brendon Wickham wrote: I can vouch for what Paul says. I started in Python 3 years ago, and I did so with a web application (still working on it!). I'm using the cgi approach, and it certainly teaches you the concepts. I fail to see how starting with a framework is a good idea if you don't know how the frameworks work (or what they're actually doing). It would be a bit like doing a web page in Dreamw***er and thinking you understand HTML/CSS. I couldn't agree less. Using the CGI module is more akin to trying to write Dreamweaver when you want to build a static website... Just use 2.6 and pick a framework of your choice. From what the OP described, Django or Pylons would fit the bill well. Chris -- Simplistix - Content Management, Batch Processing & Python Consulting - http://www.simplistix.co.uk -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
> > Anyway, for simple web programming, frameworks are not worth the > hassle. Just use the cgi module. > > I can vouch for what Paul says. I started in Python 3 years ago, and I did so with a web application (still working on it!). I'm using the cgi approach, and it certainly teaches you the concepts. I fail to see how starting with a framework is a good idea if you don't know how the frameworks work (or what they're actually doing). It would be a bit like doing a web page in Dreamw***er and thinking you understand HTML/CSS. B -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
On 25 Oct, 11:52 pm, a...@baselinedata.co.uk wrote: I am very much new to Python, and one of my first projects is a simple data-based website. I am starting with Python 3.1 (I can hear many of you shouting "don't - start with 2.6"), but as far as I can see, none of the popular python-to-web frameworks (Django, CherryPy, web.py, etc.) are Python3 compatible yet. So, what can I use to start my web programming experience using 3.1? Any help would be appreciated. don't - start with 2.6 Alan -- http://mail.python.org/mailman/listinfo/python-list -- http://mail.python.org/mailman/listinfo/python-list
Re: Web development with Python 3.1
Alan Harris-Reid writes: > I am very much new to Python, and one of my first projects is a simple > data-based website. I am starting with Python 3.1 (I can hear many of > you shouting "don't - start with 2.6"), but as far as I can see, none > of the popular python-to-web frameworks (Django, CherryPy, web.py, > etc.) are Python3 compatible yet. > > So, what can I use to start my web programming experience using 3.1? Does it occur to you that the unavailability of those frameworks is part of the REASON they say to use 2.x? Have you answered your own question? Anyway, for simple web programming, frameworks are not worth the hassle. Just use the cgi module. If you want to use a framework, well, you are the one who decided to zoom off into the 3.1 wilderness before the framework developers got there. If you're an experienced programmer in other languages and you're determined to use a framework, maybe a worthwhile Python learning project would be to help port your favorite framework to 3.1. -- http://mail.python.org/mailman/listinfo/python-list
