Generated by StatCvs v0.2-dev
Pages: << Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 Next >>
Summary Period: 2001-01-26 to 2001-01-18 (Commits 3945-3994 of 12744)
Made the following changes to inifile.cpp:
-*- Eliminated compilation warnings about implicit conversion of void* to
some other type.
-*- GetYesNo() now also recognizes "1" as true in addition to "yes",
"true", and "on".
5 lines of code changed in:
Fixed another bug in radiosity.
This time about normals when really really up close for both lumels.
Taking only one normal value would give bad results - thus, the code
now averages for the corners of all lumels....
This seems to fix the painting.
14 lines of code changed in:
fixed radiosity for the radtest level. Set source_patch_size in the config
to 1 (did not debug that yet). The 1st problem was the fact that
vectors a,b when a*b is taken (to get cosinus), lengths must be one.
a*b/ length(a)/length(b) is only a valid approximation when both a and b
are pretty large. Close to zero this fails.
The second problem was that when lumels overlap in space (i.e. two polygons
with the same texture seamlessly aligning) the distance would become almost
zero. Added a test, and at such a moment a derivative of radiosity is
applied (the diffuse*area*visibility/PI is taken, no angles/distances are
used). This seems to remove all glitches from the radtest level.
12 lines of code changed in:
Fixed a small part of the light explosion bug.
(at the cost of a little speed, must do a Normalize() for each step now.)
9 lines of code changed in:
...
2 lines of code changed in:
update
1 lines of code changed in:
...
1 lines of code changed in:
Update.
4 lines of code changed in:
working
2 lines of code changed in:
update
0 lines of code changed in:
with the exception of yuv2rgb conversion its working
83 lines of code changed in:
Update.
6 lines of code changed in:
extended PLUGINS variable for compiling opendivx codec and avi format
2 lines of code changed in:
and the work goes on
1 lines of code changed in:
a fix here and a fix there
93 lines of code changed in:
added endianess conversion
78 lines of code changed in:
Fixed a bug introduced in the last revision.
1 lines of code changed in:
Augmented cspython.mak so that it is more permissive about locating the
Python static link library.
10 lines of code changed in:
Made the following changes to csConfigFile and iConfigFileNew:
-*- Added SubsectionExists(), which is similar to KeyExists() except that
it tests if at least one key exists with the supplied prefix, rather
than testing if a particular key exists.
-*- Renamed KeyExist() to KeyExists().
-*- Added GetVFS() to complement existing GetFileName().
-*- Renamed arguments in several method prototypes to clarify and better
document their usage and intention.
-*- Upgraded and clarified header comments. Also removed incorrect
references to INI-format files from comments.
Made the following changes to csConfigFile:
-*- Now preserves blank lines in configuration files at save time instead
of folding them out. The blank lines are considered part of the
comments.
-*- Fixed bug: SaveNow() was not respecting assertion made by comments in
header that a comment could be present but empty.
-*- Now performs more thorough whitespace checking in LoadFromBuffer().
-*- Eliminated the ugly `goto' statements from LoadFromBuffer().
-*- Fixed bug: LoadFromBuffer() was not properly handling CRLF line
terminators (though it was handling LF and CR individually).
-*- LoadFromBuffer() now keeps track of the line number of the file being
parsed.
-*- LoadFromBuffer() now emits diagnostic messages if it encounters parse
errors rather than silently ignoring them.
-*- Fixed bug: LoadFromBuffer() was setting the comment on a node even if
the source file did not include a comment.
-*- The methods SetStr(), SetInt(), SetFloat(), SetBool(), and
SetComment() are now much smarter about setting the Dirty flag, and
only do so if the new value actually differs from the old.
-*- GetBool() now recognizes the string "1" as true in addition to "true",
"yes", and "on".
-*- Augmented FindNode() so that it can search for a subsection as well as
a full key. This functionality is used by the new SubsectionExists()
method.
-*- csConfigIterator::CheckSubsection() now works correctly when
enumerating the entire database (rather than just a subsection).
-*- Resolved several const vs. non-const issues. Eliminated the ugly casting
of pointers from const to non-const.
-*- Moved the csConfigNode and csConfigIterator class declarations out of
cfgfile.h and into cfgfile.cpp. These are entirely private classes,
so there was no need to publish them in a public header.
503 lines of code changed in:
further work
27 lines of code changed in:
-dummy audio stream
-further work on AVI
858 lines of code changed in:
Forgot to commit.
3 lines of code changed in:
Small update to python scripting system.
6 lines of code changed in:
Update.
16 lines of code changed in:
moved map2cs and the freetype plugin to the new config system
27 lines of code changed in:
cleaned up usage of csString in the new config system
9 lines of code changed in:
csConfigFile(iBase *pBase) was missing
1 lines of code changed in:
csConfigFile(Base *pBase) constructor was missing
6 lines of code changed in:
moved SCF to the new config system
4 lines of code changed in:
Fixed a few errors from the penultimate revision (1.35).
3 lines of code changed in:
moved SCF to the new config system
6 lines of code changed in:
moved the VFS to the new config system
101 lines of code changed in:
fixed a bug with comments in the new config system
5 lines of code changed in:
further work on video
70 lines of code changed in:
created new stream interface iStream, iAudioStream, iVideoStream, iCodec, iStreamIterator and iStreamFormat
173 lines of code changed in:
Beginning of AVI implementation (where the opendivx will live as an CODEC).
Dont try to compile :)
777 lines of code changed in:
Update.
3 lines of code changed in:
a video decoder interface
20 lines of code changed in:
added line to compile opendivx plugin (you have to uncomment it)
1 lines of code changed in:
- Removed support from sprites and lights for an old and never-used
dynamic lighting registration system. If we still need this later
we'll have to redo it another way.
- Extended iSprite3DState a bit.
- Extended iMeshObject with GetRadius() function which will return
the radius of the object in object space. Also added GetRadius()
to csMeshWrapper.
- Extended all mesh object plugins to support this function.
21 lines of code changed in:
Update.
20 lines of code changed in:
moved the glide renderer and the canvas plugins to the new config system
55 lines of code changed in:
removed the remaining entries about the Directx 5 renderer from the config
files and moved the renderers to the new config system
112 lines of code changed in:
moved all sound modules to the new config file system
25 lines of code changed in:
Removed some old COM (pre-SCF) goop from the Macintosh port.
2 lines of code changed in:
Fixed a 'delete' vs 'new[]' bug in csPolygonSet and csSprite3D.
3 lines of code changed in:
- Extended iMeshObject with HardTransform() to do a hard
transform on the object. Also added this to csMeshWrapper.
Also added SupportsHardTransform() so that an application
can test if HardTransform() is supported.
- Extended iMeshObjectFactory with the same two functions. Sometimes
HardTransform() only makes sense on the factory.
- Implemented HardTransform() in all mesh object plugins. Note
that this transformation doesn't affect the orientation of
the mesh object. Only the location is changed. Even for particle
systems and cube although it could be useful there.
Note that for spr2d, snow, and rain HardTransform() don't do
anything.
- Fixed a bug in ball plugin if some setting was changed after
already being initialized.
74 lines of code changed in:
- Fixed the bug which Martin Geisse reported with sprite frames
and actions. They were contained within a csNamedObjVector
which is not right because they are not csObject's.
- Added a 'virtual' to the destructor of the new ini file stuff
because otherwise gcc gave warnings.
61 lines of code changed in:
Update.
9 lines of code changed in:
moved the dungeon plug-in to the new config system and fixed some bugs
in both modules
2 lines of code changed in:
Pages: << Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 Next >>
Generated by StatCvs v0.2-dev