| 04:51:30 | hobu: | jmckenna: I don't like the two note links to the PDF docs. Too distracting http://mapserver.org/documentation.html |
| 04:51:31 | sigabrt: | Title: MapServer 5.6.1 Documentation MapServer 5.6.1 documentation ( at mapserver.org ) |
| 04:52:22 | jmckenna: | i'm not sure how to respond : ) |
| 04:52:36 | jmckenna: | i am against versioning personally |
| 04:52:47 | jmckenna: | it is a nightmare. |
| 04:52:53 | jmckenna: | but, suggestions |
| 04:53:01 | hobu: | PDF versioning is ok IMO. HTML, not so much. |
| 04:53:27 | jmckenna: | i will combine the 2 notes into one...is that acceptable? |
| 04:53:28 | aboudreault: | I would put that in the download page. |
| 04:53:57 | hobu: | also, aboudreault and gislars need to do their shell script magic so the en->de copying works on windows too : ) ( or suggest how to do it ) |
| 04:54:17 | jmckenna: | yes i was not able to do much on windows |
| 04:54:52 | aboudreault: | I don't know how to copy files on windows |
| 04:54:56 | aboudreault: | : ) |
| 04:55:08 | aboudreault: | gislars: !!!!!!! need help! :P |
| 04:55:10 | hobu: | IIRC, there's a copy-no-overwrite option |
| 04:56:04 | jmckenna: | i will remove the 54 link and place in downloads as suggested. |
| 04:56:19 | : | * gislars is powering on his windows machine |
| 04:56:27 | hobu: | btw, what is our documentation policy now? How do we notify de writers of updates in the en set? |
| 04:57:38 | jmckenna: | i would say that any documentation change in en should be filed as a ticket for gislars, or other language leads. |
| 04:58:02 | hobu: | our devs don't file doc tickets for us now |
| 04:58:20 | hobu: | I'm skeptical more process is going to be followed |
| 04:58:28 | jmckenna: | i see that....devs doc themselves ( which i am not a huge fan of personally ) |
| 04:58:32 | aboudreault: | Can't we use TracRSS or something of a specific directory ? |
| 04:58:41 | gislars: | additional anyone working on documentation could subscribe a rss feed |
| 04:59:40 | hobu: | which feed? I see http://trac.osgeo.org/mapserver/browser/branches/branch-5-6/docs/en doesn't really have one |
| 04:59:41 | sigabrt: | Title: /branches/branch-5-6/docs/en - MapServer - Trac ( at trac.osgeo.org ) |
| 05:00:13 | gislars: | http://trac.osgeo.org/mapserver/timeline?from=01%2F21%2F10&daysback=30&changeset=on&update=Update |
| 05:00:16 | sigabrt: | Title: Timeline - MapServer - Trac ( at trac.osgeo.org ) |
| 05:00:27 | gislars: | that lists all repository checkins |
| 05:00:56 | gislars: | maybe we could add something lice ( doc ) to the commit message |
| 05:01:02 | gislars: | so we can filter on it |
| 05:01:20 | hobu: | duh, we can make a new report. That'll generate an RSS feed |
| 05:02:20 | gislars: | ok, that would help to monitor thing, but every doc writer has to follow itslef |
| 05:02:33 | jmckenna: | nothing beats a ticket - emails, RSS feeds are forgotten...we need to trac changes, and a ticket should be filed for any documentation change required |
| 05:03:23 | gislars: | hmm really for all? |
| 05:03:24 | : | * hobu shows jmckenna the 764 unclosed tickets we already have ; ) |
| 05:03:48 | jmckenna: | it's only my opinion. |
| 05:03:55 | gislars: | if I just want to add a new translation or fix a typo i don't want to file a ticket before |
| 05:04:23 | jmckenna: | but if you want the same done to other languages, yes file a ticket |
| 05:05:24 | gislars: | so maybe tracking changes to en version would be enough? |
| 05:06:04 | gislars: | I don't plan to add something to the german version that isn't in en version |
| 05:06:06 | jmckenna: | and how are other language leads notified of their remaining tasks? |
| 05:07:29 | gislars: | hmm the always have to compre with the en docs |
| 05:08:11 | gislars: | it would be nice to have a list i could work on, but I don't see how to build such a list |
| 05:11:01 | aboudreault: | I guess RSS feed on all /en doc changes is what you need gislars. |
| 05:11:20 | hobu: | yep, can't find an easy way to do that. asking on #trac |
| 05:15:30 | gislars: | http://trac.osgeo.org/mapserver/log/trunk/docs |
| 05:15:31 | sigabrt: | Title: /trunk/docs ( log ) - MapServer - Trac ( at trac.osgeo.org ) |
| 05:15:33 | gislars: | something like this |
| 05:15:57 | gislars: | thats for /trunk/docs |
| 05:16:17 | gislars: | so each branch has its own list |
| 05:17:18 | aboudreault: | http://trac.osgeo.org/mapserver/log/trunk/docs?limit=100&mode=stop_on_copy&format=rss |
| 05:17:20 | sigabrt: | Title: MapServer: Revisions of /trunk/docsMapServerRevision 9715: updated trunk docs versionRevision 9712: fixed the init make ruleRevision 9710: ( doc ) added new target to build makefileRevision 9706: ePub supportRevision 9705: deinterlace image for rst2pdf ( PIL ) support somedayRevision 9703: import tweak to support building with Sphinx 1.0Revision 9702: add pdf builder target ( not working yet )Revision 9701: Update OL layout/css/js of the website in trunk ( at trac.osgeo.org ) |
| 05:34:55 | danmo: | hobu, jmckenna, gislars, aboudreault: thanks for your work on the /de stuff. One question though: I am no search engine expert, but I suspect the search engines may end up indexing the English pages duplicate under /de and we probably don't want that. Could we use something like <META NAME="ROBOTS" CONTENT="NOINDEX"> in those pages to prevent indexing them? |
| 05:34:59 | danmo: | http://www.robotstxt.org/meta.html |
| 05:35:00 | sigabrt: | Title: The Web Robots Pages ( at www.robotstxt.org ) |
| 05:36:30 | gislars: | good point |
| 05:36:31 | gislars: | anyone here who knows somethin about search engines? |
| 05:39:29 | gislars: | if I read this right, than we have to add the META tag to all copied files |
| 05:39:53 | danmo: | yes, I would think so |
| 05:40:02 | aboudreault: | one two split test |
| 05:40:13 | danmo: | welcome back aboudreault |
| 05:40:23 | aboudreault: | finally, crichton was in split >:| |
| 05:41:13 | danmo: | my turn to be disconnected... what's going on? |
| 05:41:51 | jmckenna: | yes its a bad IRC day it seems |
| 05:41:55 | aboudreault: | freenode is experiencing a few problems since a while |
| 05:42:22 | aboudreault: | there is a lot a irc split, might be due to DDoSer. |
| 05:42:39 | jmckenna: | aboudreault: you are right, doc pdfs make more sense in downloads.txt ( i am humbled again ) : ) |
| 05:42:51 | aboudreault: | jmckenna: woot! : ) |
| 05:42:55 | jmckenna: | : ) |
| 05:46:01 | aboudreault: | danmo: Did you explain them the search engine thing ? Did they agree to do what we said ? |
| 05:56:00 | danmo: | I wrote: thanks for your work on the /de stuff. One question though: I am no search engine expert, but I suspect the search engines may end up indexing the English pages duplicate under /de and we probably don't want that. Could we use something like <META NAME="ROBOTS" CONTENT="NOINDEX"> in those pages to prevent indexing them? |
| 05:56:08 | danmo: | only gislars responded though |
| 05:56:56 | danmo: | hobu, jmckenna: any opinion on this NOINDEX question? |
| 05:58:29 | hobu: | I agree we need to do it, but I don't know how. |
| 05:58:46 | jmckenna: | such as producing a script that writes that text into all generated HTML doc files? i don't know how either. |
| 05:59:15 | aboudreault: | I suppose it's can be done easily in bash. but no idea for windows. |
| 05:59:28 | danmo: | I was thinking of adding something to the source pages when they are copied that would prompt sphinx to either use a different header template, or insert the meta tag |
| 05:59:35 | jmckenna: | it could be added to 'build_html_docs.sh' on server |
| 05:59:37 | aboudreault: | ah... not really useful to work on windows. |
| 06:00:52 | danmo: | Does anyone really still use Windows for docs? ( Not joking ) I thought you were all either on Macs or Linux |
| 06:01:24 | aboudreault: | anyway, this is only for the officiel windows... no need to do that for the windows part. |
| 06:01:36 | aboudreault: | s/officiel/official |
| 06:01:41 | jmckenna: | i do this work on Windows...but true i could do it on mac |
| 06:01:49 | aboudreault: | s/officiel windows/official site |
| 06:03:14 | danmo: | jmckenna: well, if you are better setup to do it on Windows that's fine ( I just didn't know that was the case )... but anyway even if you were missing the meta tag on Windows that would not be a big deal I think |
| 06:04:25 | jmckenna: | agreed |
| 06:05:23 | gislars: | I think the META tag is only needed on the server, isn't it? |
| 06:05:41 | gislars: | server -> mapserver.org |
| 06:05:58 | aboudreault: | exact |
| 06:06:31 | CIA-48: | jmckenna * r9720 /branches/branch-5-6/docs/en/ ( documentation.txt download.txt ): add documentation section to download page |
| 06:06:36 | gislars: | so a bash script would be enough for that |
| 06:06:47 | jmckenna: | it could be added to 'build_html_docs.sh' on server |
| 06:07:24 | gislars: | I'm just testing the "copy task" on windows |
| 06:08:10 | jmckenna: | i will look into sphinx now to see if meta tag can be placed in source files, as danmo suggested |
| 06:08:30 | aboudreault: | it's not a sphinx thing |
| 06:08:54 | jmckenna: | i know, but sphinx generates the html no? i guess i misunderstand |
| 06:10:12 | aboudreault: | since we are copying manually the missing files in the de/ directory, we'll need to check manually all the "unversionned" files in the de directory, then add manually the meta in it ( if it's an html file ) |
| 06:10:52 | danmo: | aboudreault: are you suggesting adding it to the HTML after the fact? how about adding something to the source as you copy the files? |
| 06:12:39 | aboudreault: | no.. sorry. that's right, when we copy it, they are in ReST format. |
| 06:13:49 | aboudreault: | jmckenna: then you look into what you wanted ; ) |
| 06:14:02 | aboudreault: | s/you look/you can look |
| 06:14:43 | jmckenna: | k |
| 06:15:49 | CIA-48: | jmckenna * r9721 /trunk/docs/en/download.txt: add documentation section to download page ( for trunk ) |
| 06:23:01 | jmckenna: | rST has a 'meta' directive, i will experiment |
| 06:24:02 | CIA-48: | aboudreault * r9722 /trunk/mapserver/ ( HISTORY.TXT maptemplate.c ): Fixed memory leak related to templates ( #2996 ) |
| 06:32:01 | jmckenna: | yup, works. we need to add 2 lines to each source txt that we don't want indexed |
| 06:32:14 | jmckenna: | .. meta:: |
| 06:32:24 | jmckenna: | :ROBOTS: NOINDEX |
| 06:32:34 | danmo: | varry kewl! : ) |
| 06:32:35 | aboudreault: | cool |
| 06:32:47 | jmckenna: | sweet, i'm good for something! : ) |
| 06:34:39 | danmo: | of course you are... stick around, we still need you : ) |
| 06:35:02 | jmckenna: | : ) i'm trying to grasp if the next step is manual or not |
| 06:35:11 | danmo: | aboudreault can automate it |
| 06:36:43 | aboudreault: | ok, will do it later |
| 06:36:59 | jmckenna: | k bug me when you got time |
| 06:40:16 | danmo: | instead of |
| 06:40:16 | danmo: | cp file1 ./de/file1" |
| 06:40:16 | danmo: | use |
| 06:40:16 | danmo: | ( echo "..meta::"; echo " :ROBOTS: NOINDEX" ) | cat - file1 > ./de/file1 |
| 06:42:57 | hobu: | danmo: and on windows? ; ) |
| 06:43:40 | danmo: | windows? no meta tags on windows : ) |
| 06:43:48 | jmckenna: | : ) |
| 06:44:15 | aboudreault: | not instead, but after the cp -r, I'll identify all unversionned files with svn st | "^?" or something like that |
| 06:44:31 | : | * danmo wonders whether DOS command TYPE could be used for something similar |
| 06:44:58 | danmo: | TYPE meta_header.txt file1.txt > .\de\file1.txt |
| 06:47:08 | danmo: | aboudreault: I don't know your script so you get to make the last call... I just felt like playing a bit with bash just for fun |
| 06:48:00 | aboudreault: | Bash is missing you? : ) |
| 06:48:37 | danmo: | not that I miss bash that much, but it's more fun than writing a quote ; ) |
| 06:50:18 | nhv: | f = open( ./de/FILE1,'w' ); f.write( " :ROBOTS: NOINDEX\n" |
| 06:51:12 | nhv: | f.write( open( FILE1 ).read ) |
| 06:52:57 | jmckenna: | quotes pay the bills? : ) |
| 06:53:40 | jmckenna: | who said that? ha |
| 06:53:48 | danmo: | yup, right |
| 07:01:24 | danmo: | jmckenna: since this morning there is a new snowmobile trail that passes along our offices parking lot, apparently going to the hotel across the street... someone is trying to tease me and remind me that I have had no time yet for skidooing this winter |
| 07:07:23 | jmckenna: | danmo: ha! you're kidding? that's a sign! |
| 07:09:43 | CIA-48: | assefa * r9723 /trunk/mapserver/maplayer.c: Fix crash when drawing a map using UNIQUE function in the layer data #3271 |
| 08:48:59 | CIA-48: | jmckenna * r9724 /branches/branch-5-6/docs/en/download.txt: minor change in documentation download section |
| 08:49:35 | CIA-48: | jmckenna * r9725 /trunk/docs/en/download.txt: minor change in documentation download section |
| 09:25:05 | CIA-48: | aboudreault * r9726 /trunk/mapserver/ ( HISTORY.TXT configure configure.in mapgeos.c ): Bumped GEOS requirement to version 3.0+ ( #3215 ) |
| 09:41:33 | outer_space: | how do i know whats going on when i add a polygon layer and it shows up just blank? |
| 10:00:47 | jmckenna: | i would - remove all other layers, only have the problem layer, use ogrinfo to get extents of file, then set extent in mapfile to those values, and retry test with shp2img utility |
| 10:15:37 | CIA-48: | dmorissette * r9727 /branches/branch-5-6/docs/en/download.txt: Update with latest download links ( from 5.4 docs branch ) |
| 10:48:10 | aboudreault: | gislars: there ? |
| 11:39:46 | jmckenna: | aboudreault: ping |
| 11:40:42 | aboudreault: | jmckenna: pong |
| 11:41:30 | jmckenna: | on xblade14 ( mapserver.org ), in /osgeo/mapserver/, is nightly.sh executed every night? |
| 11:42:39 | jmckenna: | essentially I am wondering if a script that is called in that, build_clean_docs.sh, is executed every night |
| 11:43:49 | aboudreault: | who should we clean the docs ? |
| 11:44:07 | aboudreault: | I think sphinx detect all changed and rebuild if necessary |
| 11:44:34 | jmckenna: | i did not create this script, but, i would like to make sure the pdf is generated each night |
| 11:44:45 | aboudreault: | let me check |
| 11:45:32 | aboudreault: | 30 3 * * * /osgeo/mapserver/nightly.sh |
| 11:45:43 | aboudreault: | every night at 3h30AM. |
| 11:45:46 | jmckenna: | great, thanks. |
| 11:45:55 | aboudreault: | the cron jobs are in the warmerdam crontab. |
| 11:46:11 | jmckenna: | ok, i cheated and asked you...i could have dug ; ) |
| 11:47:11 | aboudreault: | hehe nah, that's ok |
| 11:47:17 | jmckenna: | thanks |
| 11:47:24 | aboudreault: | np |
| 11:47:45 | outer_space: | omg it works as expected with shp2img! |
| 11:48:17 | jmckenna: | outer_space: hmm. this means, sort of bad news ha, that your other day layers may be in a different projection |
| 11:48:59 | outer_space: | i have this file with 51 polygons for all the states |
| 11:58:50 | jmckenna: | try your layers in a desktop GIS like QGIS ( qgis.org ) and verify that they are all in the same projection |
| 12:09:35 | outer_space: | just 1 layer with 51 polygons for each state + dc |
| 12:12:42 | outer_space: | opens fine in qgis |
| 12:41:48 | hobu: | aboudreault: we need to ensure the docs are built clean every night because there are some instances where sphinx's change detection fails. |
| 12:42:22 | hobu: | and the ( old ) PDF generate was sensitive to some of the issues. I hope to do what's necessary to make the rst2pdf builder work, as it will look much better |
| 12:44:56 | aboudreault: | I see. |
| 12:46:17 | aboudreault: | a script could be done and placed in the warmerdam's cronjobs too |
| 12:48:54 | aboudreault: | non-translated de html pages have their metadata against robots now. |
| 14:22:41 | CIA-48: | dmorissette * r9728 /trunk/mapserver/mapscript/php3/php_mapscript.c: Re-add PHP4-specific code removed by accident in r9699 for ( #2170 ) |
| 17:14:19 | CIA-48: | assefa * r9729 /trunk/mapserver/ ( HISTORY.TXT maptemplate.c ): Add shplabel tag support in templates ( #3241 ) |
| 23:19:27 | leo-unglaub_work: | hi, i have a little problem with compiling mapserver on my linux server. the ./configure breaks with follow line: configure: error: Could not find gd.h or libgd.a/libgd.so in /usr/local/lib |
| 23:19:42 | leo-unglaub_work: | but the file exists in this place... |
| 23:19:56 | leo-unglaub_work: | -rw-r--r-- 1 root staff 618338 21. Jän 15:15 libgd.a |
| 23:20:10 | leo-unglaub_work: | lrwxrwxrwx 1 root staff 14 21. Jän 15:15 libgd.so -> libgd.so.2.0.0 |
| 23:20:10 | leo-unglaub_work: | lrwxrwxrwx 1 root staff 14 21. Jän 15:15 libgd.so.2 -> libgd.so.2.0.0 |
| 23:20:31 | leo-unglaub_work: | do you have any ideas? |
| 23:20:52 | racicot: | leo-unglaub_work: add /usr/local/lib to your /etc/ld.so.conf |
| 23:21:31 | racicot: | then sudo ldconfig |
| 23:22:54 | leo-unglaub_work: | i have done this yesterday if i start compiling.... |
| 23:22:59 | leo-unglaub_work: | the same error.. |
| 23:23:50 | leo-unglaub_work: | ldconfig -p shows me the same.. |
| 23:23:51 | leo-unglaub_work: | libgd.so.2 ( libc6 ) => /usr/local/lib/libgd.so.2 |
| 23:23:51 | leo-unglaub_work: | libgd.so.2 ( libc6 ) => /usr/lib/libgd.so.2 |
| 23:23:51 | leo-unglaub_work: | libgd.so ( libc6 ) => /usr/local/lib/libgd.so |
| 23:25:06 | leo-unglaub_work: | the line in my ./configure script is this one: --with-gd=/usr/local/lib \ |
| 23:27:24 | racicot: | maybe try just --with-gd and not specify path explicitly ... let the system find it |
| 23:27:49 | : | * racicot is tired and not sure... sorry |
| 23:29:19 | racicot: | and leo-unglaub_work you are sure you ran ldconfig after changing the ld.so.conf? |
| 23:29:42 | leo-unglaub_work: | jep, i run the command... |
| 23:29:52 | leo-unglaub_work: | i will try without a path... |
| 23:30:20 | leo-unglaub_work: | oh nice !!!!without a path works !!! |
| 23:30:27 | leo-unglaub_work: | thank you ! |
| 23:30:39 | racicot: | sure |
| 23:30:46 | racicot: | glad that worked |
| 23:30:59 | : | * racicot heads to bed |
| 23:32:02 | outer_space: | fajita: which channel? |
| 23:32:36 | leo-unglaub_work: | hmm, a mapserv -v works.. |
| 23:32:38 | leo-unglaub_work: | nice.. |
| 23:32:55 | leo-unglaub_work: | but if i run a map with it i get a fatal error: |
| 23:32:55 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] libpng warning: Attempted to set both read_data_fn and write_data_fn in |
| 23:32:55 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] |
| 23:32:56 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] libpng warning: the same structure. Resetting read_data_fn to NULL. |
| 23:32:56 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] |
| 23:32:57 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] gd-png: fatal libpng error: image size exceeds user limits in IHDR |
| 23:33:00 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] gd-png: EXTREMELY fatal error: jmpbuf unrecoverable; terminating. |
| 23:33:03 | leo-unglaub_work: | [Fri Jan 22 09:39:30 2010] [error] [client 212.17.123.10] Premature end of script headers: mapserv_leo |
| 23:33:06 | leo-unglaub_work: | very funny... |