source: trunk/gsdl/docs/TODO@ 1200

Last change on this file since 1200 was 1200, checked in by sjboddie, 24 years ago

* empty log message *

  • Property svn:keywords set to Author Date Id Revision
File size: 6.9 KB
Line 
1
2configuration/installation:
3
4unix:
5
6 - look into getting perl scripts to run correctly on any operating
7 system (i.e. the #! stuff) -- those in cgi-bin particularly need
8 to work on windows too (#!perl on windows, #!/use/bin/perl on
9 linux) -- perlrun manpage
10
11 - add fastcgi to distribution
12
13windows:
14
15 - use Installshield to create windows packages - wait for George to
16 decide what he's going to do about InstallShield
17
18 - call server version of cd-rom executable 'standard' and other
19 version 'backup' (or maybe 'gold' and 'silver'?)
20
21 - fix up Local Library - somehow need to have option of old server
22 and single-user (or Gold and Silver?) within same executable.
23
24
25tidy up text versions of macro files
26
27see if gimp perl scripts can be set up to handle macrons in a similar may
28to umlauts
29
30look into getting Gimp perl scripting to work on windows
31
32create default collection images at build time with gimp script
33
34fix query caching - do caching in receptionist, finish off cross collection
35searching, remove any remaining phrase-searching stuff in queryaction -
36maybe wait for mgpp before worrying about this
37
38create default images/macros for remaining Dublin core metadata elements -
39maybe have a separate macro file for all these things to make it easier to
40add to
41
42get remaining plugins tidied up and using passed in options - file
43extension option for all plugins (general option within BasPlug)
44
45get all plugins to use multiread object and throw away the specialized GB
46plugins
47
48write a gzip/bzip/zip/tar plugin and take any remaining gzip stuff out of
49rest of plugins
50
51add Stuart's acronym extraction stuff as a general plugin option (maybe
52Dana's date extraction stuff too?)
53
54tidy up classifiers, make collection-specific ones more object oriented,
55add section and compact options instead of using separate classifiers - get
56classifiers using same option passing code as plugins now use
57
58create a general classifier for doing stuff like NPepaList.pm does. should
59handle things like HDLs organization classification, HDLs magazines section
60etc.
61
62create a BasClass object for classifiers (like plugins currently
63have). probably do all this classifier stuff for gsdl-3.0 as new option
64passing stuff will require changes to all the collect.cfg files
65
66
67config files -- for gsdl-3.0
68
69 - extend configuration file syntax - tidy up all the format stuff -
70 replace DocumentImages, DocumentTitles and DocumentHeading with
71 DocumentColumn stuff (develop New_Config_Format-branch CVS branch)
72
73 - sort out arrows/paging of browserclasses (for queryaction too!)
74 DocumentBottomArrows?? - want arrows at bottom as well as top of pages
75 for gberg type collections
76
77 - tidy up automatic help text - need to work out type of collection at
78 build time - get rid of the hacky HelpNoDocs, HelpBibDocs and
79 HelpBookDocs config file entries
80
81 - fix up collection specific metadata - content negotiation??
82 "chapters" = "Kapiteln"
83 "section titles" = "Sektionstiteln"
84 "entire books" = "ganzen Büchern"
85
86 - add ability to use a separate formatstring for each index when
87 displaying query results
88
89
90append to init and error logs instead of overwriting
91
92catch niupepa up (and keep cstr up) with new development branch. hopefully
93changes to configuration files will minimize the need for specialized
94collection-specific receptionists
95
96set up cstr to use configure like niupepa
97
98fix up formattools - bug preventing nested If/Or - check that no 'here'
99pointers can run past 'end'
100
101get numwords (and maybe other stats?) working in collection info - maybe
102set as macros
103
104tidy things up so that get_collectinfo is only called once per collection
105(at init time for fastcgi) - it's currently being called all over the
106place. maybe load collection info into recpt.collectinfo structure. load in
107as required for each collection so structure would slowly fill up with all
108collections when using fastcgi
109
110get BrowseFilter EndResults option to accept -1
111
112add Language metadata to all documents by default??
113
114fix the way the interface language preference is implemented - should check
115somewhere that current language matches one of those specified in config
116file
117
118bug when classification is empty. CL number still gets incremented so
119formatstrings don't match as expected
120
121create some kind of warning when an attempt is made to build an index on
122metadata that is never set. mg currently just craps out during the build
123
124usage page (420 student doing this)
125
126add browserclasses to statusaction
127
128tidy up ohist to remove need for public_html/ohist stuff
129
130WebPlug - optimize to run reasonably quickly? - Currently works (I think)
131but processes each frame within each combination of a frameset - this means
132some html documents are processed hundreds of times and it takes forever to
133build a small collection
134
135look into possibility of clearing highlighting whenever you go to browse
136etc. i.e. highlighting only works when coming directly from a search
137
138update collections - set up auto-updating where needed
139 - csbib
140 - tidbits
141 - tcc - mail updating
142 - niupepa
143 - gberg
144
145sort documents between importing and building
146
147get polling for new collections to work when using fastcgi
148
149ipc and unesco collections have no 'about' text (maybe others too).
150
151compiler warnings on puka
152
153windows cd-rom version
154
155 - message in browser selection box to say that IE isn't compatible
156 with 'silver' version (instead of simply greying out selection of
157 IE as currently happens).
158
159 - look into Belgium bug ("Netscape was unable to create a network
160 socket connection... " -- NT 4.0 -- no internet connection -- run
161 programs/accessories/communication/internet connection wizard,
162 answer LAN to all questions -- bug appears to be present if you
163 get above error message when attempting to access 127.0.0.1 --
164 look into testing for this problem from InstallShield, maybe try
165 reading from 127.0.0.1 and see if it times out?
166
167look into need for _LITTLE_ENDIAN flag in mg
168
169tidy up 'related work' page - probably just use simple html page like for
170'publications' page. Remove related images from distribution.
171
172ohists RealAudio server appears to be broken
173
174tidy up 'standard' home page. get columns of collections to behave more
175reasonably
176
177don't put notbuilt field into build.cfg unless it's used - it causes too
178much confusion
179
180'post' data doesn't appear to work with fastcgi
181
182capability to view actual BibTeX entries of csbib (and maybe other?)
183collections
184
185scripts for installing w3mir - a .bat version of davids install.sh files
186
187sorting of unicode titles (arabic collection).
188
189get file locking to work on non GSDL_USE_IOS_H compilers (VC++ 6.0)
190
191produce an error message when no "built" collections are available
192(currently only produce an error when no collections at all are available)
193
194have another go at getting gsdl_system() function to work properly on
195windows 95 (for end-user collection building)
196
Note: See TracBrowser for help on using the repository browser.