Hi, On 18 March 2014 17:12, David Li <[email protected]> wrote: > Just some thoughts: > > I agree with Jason that ultimately, integrating a library specifically > focused on Python and web plotting is the best course for SymPy and > Gamma/Live in general. > > Gamma/Live's problem is that it runs on App Engine - anything that requires > a server won't work. (I think Bokeh/WebAgg fall into this category, but I'm > not sure.
Bokeh can just generate JSON/JavaScript/HTML, so no server is required (thought possible as well). I'm not sure if it would run on App Engine anyway, because of its dependencies (numpy, pandas, etc.), but we will make them optional (most of Bokeh's dependencies are used by the bokeh.server module anyway). Mateusz > WebAgg also isn't available on App Engine yet anyways.) > > WebAgg seems to render the plot on the server and send frames to the client. > Locally it ran at an acceptable (but slow) rate; I'm not sure how well this > would hold up on App Engine. > > One of the reasons that Gamma uses its own plotting is because it lets the > user pan the plot and load more of the plot on demand. > > On mpld3: it would be nice to integrate just to support more plots, but for > 3D specifically it won't help. > On passing JSON data to the browser to render: this is what we do already. > > David > > On Tuesday, March 18, 2014 8:47:50 AM UTC-7, SAHIL SHEKHAWAT wrote: >> >> That exactly what i was looking for! >> I think i will go by this concept only...its awesome! >> about using Json mathics guys also use json to render 3D plotting for >> Three.js, Its good enough and works. >> Thanks again. :) >> >> >> On Tue, Mar 18, 2014 at 8:52 PM, Jason Moore <[email protected]> wrote: >>> >>> https://github.com/pydy/pydy-viz >>> >>> We basically generate data on the python side, then write a json file to >>> disk, then we have a generic html/js/css template that uses three.js to load >>> in all the data from the json file and create a scene. The json file is the >>> communication between python and js. This is rudimentary, but works and I >>> think it would work for your idea too. >>> >>> >>> Jason >>> moorepants.info >>> +01 530-601-9791 >>> >>> >>> On Tue, Mar 18, 2014 at 11:19 AM, SAHIL SHEKHAWAT <[email protected]> >>> wrote: >>>> >>>> can you please point me out the direction where i can learn how PyDy >>>> works? >>>> >>>> >>>> On Tue, Mar 18, 2014 at 8:28 PM, Jason Moore <[email protected]> wrote: >>>>> >>>>> But even if you implement something with Mathbox.js then it would be >>>>> nice if users can use it locally or through Live/Gamma. That is the >>>>> approach >>>>> with took with pydy-viz. Right now we can run them locally and if you want >>>>> embed it in a website you can. >>>>> >>>>> >>>>> Jason >>>>> moorepants.info >>>>> +01 530-601-9791 >>>>> >>>>> >>>>> On Tue, Mar 18, 2014 at 10:57 AM, Jason Moore <[email protected]> >>>>> wrote: >>>>>> >>>>>> If we already use matplotlib for 3D plotting, then a long term >>>>>> solution to this is to add three.js to the web backend of any one of the >>>>>> new >>>>>> plotting libraries that are popping up (bokeh, mpld3, matplotlib's webagg >>>>>> backend). Then you can simply use the library for the backend to the >>>>>> matplotlib calls. Now, that route is more arduous and may not be the >>>>>> fastest >>>>>> course to having 3D plotting in Gamma. You can certainly implement a >>>>>> custom >>>>>> 3D web based plotter for SymPy Gamma that only works for SymPy Gamma, >>>>>> but as >>>>>> soon as these aforementioned libraries mature we will most certainly use >>>>>> them as the backend for SymPy plotting. I'm not involved in the SymPy >>>>>> plotting package so I don't know the roadmap for it or what people really >>>>>> want. But people that are doing plotting want what these new libraries >>>>>> are >>>>>> creating and one or more of them will become a de facto web plotting >>>>>> library >>>>>> for 3D graphics. But also having a working 3D plot in Gamma by the end of >>>>>> the summer will be cool, regardless if we use the more established libs >>>>>> in >>>>>> the future. If you want to go the fast and dirty route, then the >>>>>> Mathbox.js >>>>>> looks like a good choice. >>>>>> >>>>>> >>>>>> Jason >>>>>> moorepants.info >>>>>> +01 530-601-9791 >>>>>> >>>>>> >>>>>> On Tue, Mar 18, 2014 at 10:32 AM, SAHIL SHEKHAWAT >>>>>> <[email protected]> wrote: >>>>>>> >>>>>>> But we do support 3D plotting by matplotlib and also you mentioned >>>>>>> that a lot of people are working on making it browser compatible. >>>>>>> Sorry if i am irritating but i just want to clarify. >>>>>>> >>>>>>> >>>>>>> On Tue, Mar 18, 2014 at 7:55 PM, Jason Moore <[email protected]> >>>>>>> wrote: >>>>>>>> >>>>>>>> My suggestion on melange was just to mention that if you only focus >>>>>>>> on 3D plotting for SymPy Gamma then we are missing the majority of >>>>>>>> SymPy >>>>>>>> users. My hunch is that 99% of people who use SymPy use it locally, >>>>>>>> not from >>>>>>>> SymPy live or SymPy Gamma. If we currently don't support 3D plotting >>>>>>>> locally, then the effort may be more fruitful adding 3D plotting to >>>>>>>> SymPy >>>>>>>> for local use. If 3D plotting for SymPy were implemented in such a way >>>>>>>> that >>>>>>>> it is web compatible, then the plotting for SymPy Live and SymPy Gamma >>>>>>>> will >>>>>>>> be gained almost for "free". >>>>>>>> >>>>>>>> >>>>>>>> Jason >>>>>>>> moorepants.info >>>>>>>> +01 530-601-9791 >>>>>>>> >>>>>>>> >>>>>>>> On Tue, Mar 18, 2014 at 9:15 AM, SAHIL SHEKHAWAT >>>>>>>> <[email protected]> wrote: >>>>>>>>> >>>>>>>>> sorry for the slang but if a lot of people are working on the >>>>>>>>> similar project then what is there that i can work on! >>>>>>>>> I talked to Ondrej and David and they said that implementing 3D >>>>>>>>> plotting for Gamma is fine. Please suggest something. >>>>>>>>> >>>>>>>>> >>>>>>>>> On Tue, Mar 18, 2014 at 6:32 PM, SAHIL SHEKHAWAT >>>>>>>>> <[email protected]> wrote: >>>>>>>>>> >>>>>>>>>> So, how can i improve sympy then? >>>>>>>>>> I mean you said that we want 3D plotting for SymPy as a whole. so >>>>>>>>>> , how can i improve that? >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Tue, Mar 18, 2014 at 6:17 PM, Jason Moore <[email protected]> >>>>>>>>>> wrote: >>>>>>>>>>> >>>>>>>>>>> There are a lot people working on plotting in the browser, many >>>>>>>>>>> of them tied to python execution. There probably isn't much reason >>>>>>>>>>> to >>>>>>>>>>> reinvent the wheel here. This is another more simple project that >>>>>>>>>>> implements >>>>>>>>>>> a d3.js backend for matplotlib: mpld3.github.io . Also there >>>>>>>>>>> matplotlib has >>>>>>>>>>> a webagg backend. >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> Jason >>>>>>>>>>> moorepants.info >>>>>>>>>>> +01 530-601-9791 >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> On Tue, Mar 18, 2014 at 7:41 AM, SAHIL SHEKHAWAT >>>>>>>>>>> <[email protected]> wrote: >>>>>>>>>>>> >>>>>>>>>>>> Indeed they are and for that there is already an option of >>>>>>>>>>>> downloading from matplotlib's support. >>>>>>>>>>>> We can give an option to view the plot in the web browser and >>>>>>>>>>>> Live and Gamma are always there to help us. They can render the >>>>>>>>>>>> plots very >>>>>>>>>>>> nicely. >>>>>>>>>>>> I think it is very significant and should be my proposal for >>>>>>>>>>>> GSOC. >>>>>>>>>>>> Thanks a lot Christophe for your help. >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> On Tuesday, March 18, 2014 4:57:50 PM UTC+5:30, Christophe Bal >>>>>>>>>>>> wrote: >>>>>>>>>>>>> >>>>>>>>>>>>> Do not forget that "fixed" pictures are useful for paper >>>>>>>>>>>>> reports. On the other hand, I think that using web browsers for >>>>>>>>>>>>> sliders is >>>>>>>>>>>>> the better solution to use actually because of the dynamical >>>>>>>>>>>>> features and >>>>>>>>>>>>> the JS library that helps to build such sliders. >>>>>>>>>>>> >>>>>>>>>>>> -- >>>>>>>>>>>> You received this message because you are subscribed to the >>>>>>>>>>>> Google Groups "sympy" group. >>>>>>>>>>>> To unsubscribe from this group and stop receiving emails from >>>>>>>>>>>> it, send an email to [email protected]. >>>>>>>>>>>> To post to this group, send email to [email protected]. >>>>>>>>>>>> >>>>>>>>>>>> Visit this group at http://groups.google.com/group/sympy. >>>>>>>>>>>> To view this discussion on the web visit >>>>>>>>>>>> https://groups.google.com/d/msgid/sympy/470f4413-08ad-4382-b349-179d35ff3914%40googlegroups.com. >>>>>>>>>>>> >>>>>>>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> -- >>>>>>>>>>> You received this message because you are subscribed to the >>>>>>>>>>> Google Groups "sympy" group. >>>>>>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>>>>>> send an email to [email protected]. >>>>>>>>>>> To post to this group, send email to [email protected]. >>>>>>>>>>> >>>>>>>>>>> Visit this group at http://groups.google.com/group/sympy. >>>>>>>>>>> To view this discussion on the web visit >>>>>>>>>>> https://groups.google.com/d/msgid/sympy/CAP7f1Ag6G_Hoxq6GqkSnMPVD8h53s2bzWHckoCURY0oMRsEFJA%40mail.gmail.com. >>>>>>>>>>> >>>>>>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>> -- >>>>>>>>> You received this message because you are subscribed to the Google >>>>>>>>> Groups "sympy" group. >>>>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>>>> send an email to [email protected]. >>>>>>>>> To post to this group, send email to [email protected]. >>>>>>>>> >>>>>>>>> Visit this group at http://groups.google.com/group/sympy. >>>>>>>>> To view this discussion on the web visit >>>>>>>>> https://groups.google.com/d/msgid/sympy/CA%2BOR%3DbhVfJc9A3PiTdfb-HcEtEj5GmMTTR0x1B%3DAi%2ByK3idDjQ%40mail.gmail.com. >>>>>>>>> >>>>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>>> >>>>>>>> >>>>>>>> -- >>>>>>>> You received this message because you are subscribed to the Google >>>>>>>> Groups "sympy" group. >>>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>>> send an email to [email protected]. >>>>>>>> To post to this group, send email to [email protected]. >>>>>>>> >>>>>>>> Visit this group at http://groups.google.com/group/sympy. >>>>>>>> To view this discussion on the web visit >>>>>>>> https://groups.google.com/d/msgid/sympy/CAP7f1Ah6sPERxrh1CF-nbJa5pq%3DRQrq4K0%3DzHUjkN4mo8%2BKHmg%40mail.gmail.com. >>>>>>>> >>>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>> >>>>>>> >>>>>>> -- >>>>>>> You received this message because you are subscribed to the Google >>>>>>> Groups "sympy" group. >>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>> send an email to [email protected]. >>>>>>> To post to this group, send email to [email protected]. >>>>>>> >>>>>>> Visit this group at http://groups.google.com/group/sympy. >>>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/d/msgid/sympy/CA%2BOR%3Dbh5HE0XqMpGp1g7FSGpFxMDfo15Ff%2BeW8Rdg4k465GC%2BQ%40mail.gmail.com. >>>>>>> >>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>> >>>>>> >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "sympy" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> To post to this group, send email to [email protected]. >>>>> >>>>> Visit this group at http://groups.google.com/group/sympy. >>>>> To view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/sympy/CAP7f1AiqGXomefk2gsAcA3vk%2Bw%2Bakf-12XdZN-WfBbUdtUfcMA%40mail.gmail.com. >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>> >>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "sympy" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to [email protected]. >>>> To post to this group, send email to [email protected]. >>>> >>>> Visit this group at http://groups.google.com/group/sympy. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/sympy/CA%2BOR%3DbiD1cTBF8Z_echYzTH1aPDwxeFvsSdLbWBFejaVh3Sr-A%40mail.gmail.com. >>>> >>>> For more options, visit https://groups.google.com/d/optout. >>> >>> >>> -- >>> You received this message because you are subscribed to the Google Groups >>> "sympy" group. >>> To unsubscribe from this group and stop receiving emails from it, send an >>> email to [email protected]. >>> To post to this group, send email to [email protected]. >>> >>> Visit this group at http://groups.google.com/group/sympy. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/sympy/CAP7f1Ah%3DZEyP0hBzevepd-SoMGvoJb-qQ39DtH24jq589BmYSg%40mail.gmail.com. >>> >>> For more options, visit https://groups.google.com/d/optout. >> >> > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sympy. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sympy/639b6471-703e-4905-b2b2-f21ca2afad1b%40googlegroups.com. > > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAGBZUCaoD4XxsJrpzeVgPecDiZF5f7Cj%3Dgq35vsRjqCXgqvweQ%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
