| 05:12:11 |
: |
* jmckenna manually regenerating live docs...steve made some minor style changes |
| 05:25:17 |
hobu: |
aboudreault: re php 4. at what point do you guys make a push for getting on the swig bandwagon by building up swig's php support to the level you need for mapserver? |
| 05:26:00 |
hobu: |
it seems quite difficult to keep maintaining a 16900 line c file by hand : ) |
| 05:27:08 |
hobu: |
the php driver in swig has been getting a lot of love lately, and it purports to support up to php 6.3 |
| 05:27:24 |
crschmidt: |
... there's a php 6 now? |
| 05:31:17 |
aboudreault: |
hobu: the problem with swig is the backwards compatibility |
| 05:31:32 |
: |
* hobu meant 5.3 |
| 05:31:47 |
hobu: |
if you're ditching 4, doesn't that cure the problem? |
| 05:32:02 |
aboudreault: |
We have to check in detail if we could support the exact same API with swig, if so... we'll of course use swig |
| 05:32:24 |
crschmidt: |
hobu: I guess he means Mapscript backwards compatibility |
| 05:32:41 |
hobu: |
ah. yes, there's quite a bit of divergence |
| 05:32:55 |
aboudreault: |
yes, Mapscript backwards compatibility. function names, constructors properties.... |
| 05:45:44 |
aboudreault: |
hobu: I'm going to take some time to check what could be done with swig.... It would be definitively easier to maintain. |
| 05:46:03 |
hobu: |
ok, just thought i'd bring it up again. |
| 05:46:19 |
hobu: |
it's not something you're going to be able to do in a week though |
| 05:46:30 |
hobu: |
but maybe we can find someone who wants it bad enough |
| 05:51:09 |
jmckenna: |
aboudreault: do you object if i re-order the PHPMapScript class objects in the documentation to be alphabetical? |
| 05:51:20 |
jmckenna: |
it's getting quite hairy |
| 05:52:24 |
jmckenna: |
anyone have any objections to that? |
| 05:52:33 |
aboudreault: |
absolutely not. feel free : ) |
| 05:52:39 |
jmckenna: |
sweet thanks |
| 06:13:37 |
jmckenna: |
( speaking to PHPMapScript devs in the room ) are there set guidelines for naming of Class Objects? I see some with lowercase, some most not... can we clean this up? Should I file a ticket? |
| 06:14:22 |
jmckenna: |
maybe guidelines should be set in the PHPMapScript wiki? |
| 06:18:48 |
jmckenna: |
just a comment, not the end of the world. |
| 07:06:38 |
aboudreault: |
jmckenna: I'll follow a guideline if I rewrite the php ext. I'll probably try to be similar to swig classes. "mapObj" |
| 07:07:10 |
jmckenna: |
thanks, that would be really helpful to users. thanks |
| 07:08:18 |
aboudreault: |
though I don't really like this and that I would rather use something like myMap = new Mapserver.Map( ).... , it's important to preserve a guideline for all mapscripts languages. |
| 07:09:19 |
jmckenna: |
I mean I am happy as long as each class name follows the same guideline. |
| 07:09:57 |
jmckenna: |
good point, we should use the same swig guidelines ? ) |
| 07:10:02 |
jmckenna: |
( ? ) |
| 07:11:21 |
aboudreault: |
since swig is currently the bindings generator of ~5 other languages... we don't really have choice ; ) |
| 07:11:57 |
jmckenna: |
ok, i think i follow you ; ) |
| 07:13:09 |
aboudreault: |
"you think you follow me" because I am not clear ? or because my english sucks? lol |
| 07:14:36 |
jmckenna: |
I am not fully familiar with swig, that's all - but i do understand what it is. |
| 07:15:29 |
aboudreault: |
Ok, not very familiar with it neither. I'll learn it if I have to. |
| 07:30:28 |
hobu: |
swig: you can't do worse, but you couldn't do better |
| 07:51:15 |
CIA-48: |
jmckenna * r9761 /trunk/docs/en/mapscript/php/index.txt: general cleanup of doc, through ticket:1794, including reordering classes and adding inline links to referred classes |
| 08:07:09 |
: |
* jmckenna willdo more cleanup of doc..so that class object names use same case structure |
| 08:09:20 |
crschmidt: |
oh php |
| 08:09:23 |
crschmidt: |
and your case insensitivity |
| 08:10:03 |
: |
* crschmidt wonders what happens if you have a group and layer with the same name in MapServer |
| 08:10:38 |
hobu: |
somewhat undefined. I use that as a cheat for having multiple scaled layers sometimes |
| 08:17:59 |
CIA-48: |
jmckenna * r9762 /trunk/docs/en/mapscript/php/index.txt: more attempts are standardizing class object names in doc |
| 10:32:47 |
: |
* jmckenna likes the apache mod_fcgid changes - hosted by apache foundation, instead of separate project ===> less configuration |
| 10:33:26 |
: |
* jmckenna ( used to have to set about 10 env variables, on windows ) |
| 10:33:38 |
: |
* jmckenna buh-bye la |
| 11:45:04 |
danmo: |
jmckenna: about r9672, I believe the unwritten agreed convention for the class names was "className" and not "ClassName" |
| 11:45:29 |
danmo: |
This is what was used in the SWIG MapScript docs as well: http://mapserver.org/mapscript/mapscript.html#mapscript-classes |
| 11:45:30 |
sigabrt: |
Title: SWIG MapScript API Reference MapServer 5.6.1 documentation ( at mapserver.org ) |
| 11:49:03 |
jmckenna: |
yes i noticed that document. however throughout the php mapscript documentation ( methods etc ) we interchange that convention often, so I am not sure what to do |
| 11:49:29 |
jmckenna: |
i.e. it's in the code, so not sure how to proceed |
| 11:49:45 |
danmo: |
in the code? |
| 11:50:06 |
aboudreault: |
what classes ? |
| 11:50:11 |
jmckenna: |
method name, etc etc |
| 11:50:29 |
danmo: |
method names should also be lowercase first, e.g. getExpression( ) |
| 11:50:46 |
danmo: |
if there are instances of GetSomething( ) then they are the ones that need to be fixed |
| 11:50:55 |
danmo: |
anyway that's how things used to be |
| 11:51:01 |
jmckenna: |
yes that is why i was mentioning this earlier today |
| 11:51:05 |
aboudreault: |
since php is case insensitive, let it as it is... we'll see when we rewrite that stuff |
| 11:51:14 |
jmckenna: |
yeah agreed |
| 11:51:17 |
danmo: |
aboudreault: code != docs |
| 11:51:31 |
danmo: |
the code has to be all lowercase internally ( PHP requirement ) |
| 11:51:33 |
jmckenna: |
code for me meant in the c code |
| 11:51:42 |
danmo: |
but for docs we use className and getSomething( ) |
| 11:52:26 |
aboudreault: |
php requirement ? do you mean php3 requirement? :P |
| 11:53:00 |
danmo: |
aboudreault: no, when you register a function in a C module, you have to register the name lowercase... then in PHP code ( in scripts ) it is case insensitive |
| 11:53:01 |
jmckenna: |
the doc is littered with different conventions. i don't think anyone understands me. |
| 11:53:29 |
danmo: |
jmckenna: my vote would be to use className and getSomething( ) in docs and not worry about the C source |
| 11:53:41 |
danmo: |
no, I don't understand the issue with the source |
| 11:55:19 |
aboudreault: |
danmo: kk and if we register it with a uppercase letter, it is not case insensitive in php, that's it? |
| 11:55:31 |
danmo: |
with uppercase it just won't work |
| 11:56:02 |
danmo: |
I suspect the PHP parser lowercases everything |
| 11:56:18 |
danmo: |
that's how it used to be in PHP4 anyway. I'd be surprised if that changed in PHP5 |
| 11:56:36 |
jmckenna: |
the latest version of the doc can be viewed at http://mapserver.org/trunk/en/mapscript/php/index.html |
| 11:56:37 |
sigabrt: |
Title: PHP MapScript MapServer 5.6.1 documentation ( at mapserver.org ) |
| 11:56:42 |
aboudreault: |
not sure to understand you, there is a log a c functions registered with uppercase. |
| 11:56:58 |
jmckenna: |
if we all think it is sufficient i am ok to accept : ) |
| 11:57:27 |
jmckenna: |
but, i did voice my concerns about the different conventions used...if readers can understand then i am ok...but i was confused |
| 11:57:27 |
danmo: |
jmckenna: I see that you asked the question here this morning, but I don't think it got noticed |
| 11:57:51 |
: |
* danmo hates when decisions are made only on IRC and not broadcasted on -dev list |
| 11:58:19 |
aboudreault: |
who takes a decision? |
| 11:59:07 |
danmo: |
the PSC should decide/advise on conventions like this |
| 11:59:41 |
jmckenna: |
agreed. hmm |
| 12:00:30 |
danmo: |
jmckenna: not sure what you mean by "sufficient"? |
| 12:01:29 |
jmckenna: |
i guess i can be just causing noise - if it ain't broke don't fix it? maybe i am only one noticing different conventions in doc. is doc acceptable now in trunk? |
| 12:02:11 |
jmckenna: |
to me i see method names and objects etc using various naming conventions..i think that is the issue |
| 12:02:15 |
danmo: |
I see some inconsistent ones such as the ms_io* functions |
| 12:02:29 |
jmckenna: |
yep i see that |
| 12:02:37 |
danmo: |
I agree with cleaning things up... just don agree with the convention you chose |
| 12:03:05 |
danmo: |
i/e/ we've always className and getSomething( ) ... so I think we should continue in that direction |
| 12:03:15 |
danmo: |
and fix whatever doesn't match that |
| 12:03:40 |
danmo: |
note that those starting with ms_* are not class methods but raw functions, hence the ms_ prefix |
| 12:03:51 |
jmckenna: |
to go back to how i started this morning - maybe we should document a convention in the PHP MapScript wiki? because, truth is i do not document these methods usually |
| 12:04:20 |
danmo: |
sure, but that should not be limited to PHP MapScript, that should be for all MapScript versions |
| 12:04:30 |
jmckenna: |
we need to all agree on something..kind of pointless for me to change existing doc if we have no convention document that devs follow |
| 12:04:38 |
jmckenna: |
true true |
| 12:04:41 |
danmo: |
agreed |
| 12:04:54 |
danmo: |
that's why I say this needs to be discussed on -dev, not here |
| 12:05:07 |
jmckenna: |
i'm nodding my head - agreed |
| 12:05:14 |
jmckenna: |
ok i will send a dev email |
| 12:05:17 |
danmo: |
all that being said, I think there are higher priority docs tickets than cleaning up conventions |
| 12:05:34 |
jmckenna: |
it's just something i noticed. true i agree |
| 12:05:48 |
danmo: |
sorry if that sounds rude, but devs document stuff themsleves because docs tickets sit in Trac for too long |
| 12:10:35 |
danmo: |
aboudreault: to undertsand what I meant by "method names must be registered lowercase", look at http://pastebin.ca/1767066 |
| 12:10:36 |
sigabrt: |
Title: pastebin - Anonymous - post number 1767066 ( at pastebin.ca ) |
| 12:11:06 |
danmo: |
the C function may have upper/lowercase since it's compiled, but the registered name ( inside double quotes ) must be all lowercase |
| 12:12:08 |
aboudreault: |
danmo: yeah. but looks like we can register them using uppercase in zend engine 2. |
| 12:12:26 |
danmo: |
maybe that changed then |
| 12:13:23 |
jmckenna: |
sorry for the can of worms guys...it's just something i noticed. the doc is good and no one else mentioned it so we can move to more pressing issues. |
| 12:14:06 |
jmckenna: |
10 year anniversary of PHP MapScript? : ) |
| 12:14:08 |
danmo: |
Don't get me wrong, I do agree that following a convention is important... and we have an implicit one: className and getSomething( ) |
| 12:14:23 |
danmo: |
yeah, that must be about now |
| 12:14:33 |
jmckenna: |
wow, quite amazing |
| 12:15:11 |
aboudreault: |
danmo: probably, just notice that my last commit on php/mapscript have an uppercase in the function registration: http://trac.osgeo.org/mapserver/changeset/9744 |
| 12:15:14 |
sigabrt: |
Title: Changeset 9744 - MapServer - Trac ( at trac.osgeo.org ) |
| 12:18:35 |
danmo: |
Almost exactly 10 years ago I found MapServer and sent this to the list: http://lists.osgeo.org/pipermail/mapserver-users/2000-January/027641.html |
| 12:18:36 |
sigabrt: |
Title: Wow! ( at lists.osgeo.org ) |
| 12:18:42 |
danmo: |
That was Jan 11, 2000 |
| 12:19:22 |
danmo: |
Oh, and the first post on PHP MapScript was Jan 26th, exactly 10 years ago: http://lists.osgeo.org/pipermail/mapserver-users/2000-January/027672.html |
| 12:19:23 |
sigabrt: |
Title: PHP module ( at lists.osgeo.org ) |
| 12:19:38 |
jmckenna: |
i'm psychic!!!!!!!!!!!!!!!! freaky |
| 12:19:52 |
aboudreault: |
lol |
| 12:20:34 |
jmckenna: |
l like that first email from danmo, that's some history there |
| 12:21:15 |
jmckenna: |
a very polite user : ) |
| 12:23:19 |
crschmidt: |
wow |
| 12:24:59 |
jmckenna: |
danmo: permission to tweet? |
| 12:25:05 |
jmckenna: |
ha, im laughing |
| 12:25:07 |
danmo: |
am about to tweet |
| 12:25:16 |
danmo: |
you can retweet ; ) |
| 12:25:21 |
jmckenna: |
you got the honor...it's all yours. deal |
| 12:28:34 |
danmo: |
thx |
| 12:44:49 |
racicot: |
My hand gets gently slapped by nhv on my first mapserver mailing list post... those where the days ;- ) 2003 http://lists.osgeo.org/pipermail/mapserver-users/2003-February/040092.html |
| 12:44:50 |
sigabrt: |
Title: [Mapserver-users] New User - Compile problems... ( at lists.osgeo.org ) |
| 12:45:42 |
jmckenna: |
not plain text, ouch, hard slap! : ) |
| 12:49:23 |
crschmidt: |
wow, that's way more information than nhv usually gives these days : ) |
| 12:59:20 |
: |
* nhv has learned to enjoy parrots :- ) |
| 12:59:56 |
crschmidt: |
: ) |
| 20:26:39 |
outer_space: |
is there a file name size limit for map files? |
| 20:26:45 |
outer_space: |
for shape files? |