Guys, I have made the first part of the `bug_hunter' script that can accurately scan the whole C sources for protos, similar to what Jesus attempted.
It also detects automatically the proto imperfections retrieved from C sources blaming thing like variables without types, extra comas and funny type names/typoS :) etc. It seems to be quite precise on the things, while is still a little bit to configure. I am going to add the docs scanner and the Source Code scanner for various comparisons. Should I commit my code to the phpdoc/scripts at this point or later? Also, Ilia, I haven't heard from you yet about the work you have done regarding something similar to this bug hunter. -- Maxim Maletsky [EMAIL PROTECTED] Maxim Maletsky <[EMAIL PROTECTED]> wrote... : > > oh, I noticed :) > > Update: for now I only composed the script (partially based on Jesus's > and Brad's code) that scans the the C sources and funcsummary.txt to a > structured array of data. This, would then be compared to the XML in the docs > and the C sources to get the actual parameter counts, return types and > proto definitions. Comparing the two, you get some very interesting > results. > > This script cannot be precise as the C code is a lot tricky and i am not > going to write a C parser here. At most, it catched some 60-70% of the > C sources. But so far, in just a few hours stubbing the code, I noticed > how many inconsistencies were there. > > Let's see what Ilia is playing with. If same thing then we might want to > merge the efforts somehow, otherwise I can commit my file once I make it > be something more finalized and useful. > > > > -- > Maxim Maletsky > [EMAIL PROTECTED] > > > On Mon, 3 Feb 2003 18:00:15 +0000 (GMT) Philip Olson <[EMAIL PROTECTED]> wrote: > > > Sorry, I wrote a wrong address for Ilia so here's a > > correct one. > > > > Philip > > > > > > ---------- Forwarded message ---------- > > Date: Mon, 3 Feb 2003 17:46:08 +0000 (GMT) > > From: Philip Olson <[EMAIL PROTECTED]> > > To: Maxim Maletsky <[EMAIL PROTECTED]> > > Cc: Friedhelm Betz <[EMAIL PROTECTED]>, Gabor Hojtsy <[EMAIL PROTECTED]>, > > [EMAIL PROTECTED], [EMAIL PROTECTED] > > Subject: Re: [PHP-DOC] proto-skript [was authors....] > > > > On Mon, 3 Feb 2003, Maxim Maletsky wrote: > > > Thanks, I think this can be useful to me. > > > > > > Actually, in base of this (and others out there) script(s), I am going > > > to attempt creating a script that parses both the C code and the docs > > > tree to look for any inconsistencies. > > > > > > Somewhat like a doc-bug hunter. The goal in this case will be to > > > analyze the current documentation and get listed the functions that do > > > not correspond to the current documentation, such as in our case some > > > return values and parameter counts. Would be helpful getting a > > > machine-generated summary of incorrectly documented protos and functions > > > (like for things that have been changed since they were documented etc). > > > > > > I already saw things doing similar to that, so I've got the base. > > > > > > Any thoughts/complains/suggestions? If what comes out will work well for > > > me and nobody objects, I will then commit it to the /phpdoc/scripts. > > > > Ilia has already created something that does this and > > he's been going through it doing minor tweaks here and > > there for awhile now. You may want to talk to him. > > > > Regards, > > Philip > > > > > > > > > > > > > Friedhelm Betz <[EMAIL PROTECTED]> wrote... : > > > > > > > > > I have heard some rumors about someone creating a proto check script to > > > > > > check whether the protos in phpdoc reflect the current protos of > > > > > > functions in the php source. Take in account undocumented new > > > > > > parameters, etc. too... I don't think that anybody can get on this to do > > > > > > it manually in any reasonable time... > > > > > > > > > > Would be nice to investigate on this deeper and see whether we can do > > > > > something about it. I could go ahead to write a similar script myself. > > > > > > > > Take a look at /phpdoc/scripts/xml_proto.php > > > > never tried, but may be a good starting point. > > > > > > > > Friedhelm > > > > > > > > > > > > -- > > > > PHP Documentation Mailing List (http://www.php.net/) > > > > To unsubscribe, visit: http://www.php.net/unsub.php > > > > > > > > > > > > > -- > > > PHP Documentation Mailing List (http://www.php.net/) > > > To unsubscribe, visit: http://www.php.net/unsub.php > > > > > > > > > -- > > PHP Documentation Mailing List (http://www.php.net/) > > To unsubscribe, visit: http://www.php.net/unsub.php > > > > > > -- > > PHP Documentation Mailing List (http://www.php.net/) > > To unsubscribe, visit: http://www.php.net/unsub.php > > > > > -- > PHP Documentation Mailing List (http://www.php.net/) > To unsubscribe, visit: http://www.php.net/unsub.php > -- PHP Documentation Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
