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: 2000-10-09 to 2000-10-02 (Commits 3545-3594 of 12744)
-*- Added csHashCompute(char const*) to hashmap.{h|cpp} which quickly and
effectively computes a hash key from a string.
-*- Eliminated the bogus and broken GetReducedHash() which Brandon added
to csMD5. Brandon was using this as a bizarre form of very expensive
string hashing. To say that this was overkill is an understatement.
It was like using a sledge-hammer to kill an ant. Furthermore,
GetReducedHash() was buggy and failed to be endian-correct.
-*- Modified Brandon's motion.cpp to use the inexpensive csHashCompute()
rather than the overkill and bogus GetReducedHash(). Also reformatted
motion.cpp to follow CS formatting guidelines. (Brandon, please use
tab size 8!)
-*- Fixed a memory corruption bug in motion.cpp. Was writing past the end
of an allocated buffer when copying a string.
-*- Removed garbage which brandon left in motion.mak.
-*- Added Motion plugin to user.mak and static.cpp since Brandon forgot to
do so.
-*- Removed one of the CS_ASSERT() calls which Brandon inserted into
csBasicVector since it fundamentally changed the behaviour of the
class, even though he neglected to modify the rest of CS to take this
fundamental behavioral change into account. (Brandon, if you make
fundamental behavioral changes like this, then you must also fix every
place in the project which is affected by such changes. Do not expect
other people to keep cleaning up after you.)
221 lines of code changed in:
Further work on leafy BSP generator in levtool. Added a 'bsp'
button to test it. Select a sector from the list below and it
will take the static thing (if present, else the sector polygons)
to make a bsp. Not useful yet at the moment.
4 lines of code changed in:
Update.
20 lines of code changed in:
Some cleanup.
15 lines of code changed in:
Frame index bug that caused a crash
8 lines of code changed in:
Fixed compile errors
10 lines of code changed in:
Had a revelation and restructured the motion API
137 lines of code changed in:
Adding the motion interface to the CVS tree. The SCF interfaces & the
loader are 90% done, but the plugin and rendering loop still need to be
implemented.
182 lines of code changed in:
removed some warnings
12 lines of code changed in:
added a makro for safe ADD_REF and DEC_REF
7 lines of code changed in:
Further work on leafy BSP generator.
5 lines of code changed in:
Fixed compile error after Sunshine update to:
csFontDef* csDefaultFontServer::ReadFntFile(const char *file)
Moved 'span' decleration and init to head of function.
Hope is OK?
2 lines of code changed in:
Added a check for empty scf.cfg file so that it won't bomb out.
11 lines of code changed in:
- Started work on leafy BSP builder in levtool. This will eventually
be used for PVS.
- Fixed a bad bug in lighting. It was doing delete [] data
instead of data->DecRef ().
5 lines of code changed in:
Made the following changes to the csFont plugin:
-*- Fixed bug: Was leaking a VFS reference.
-*- Fixed bug: The destructor was leaking every csFontDef record added by
LoadFont(). The contents of the records were being freed, but the
record themselves were not.
-*- Fixed a really significant number of memory leaks in ReadFntFile()
which would occur on error conditions. It's currently using a 'goto'
(yuck!) to handle such conditions. In the future, this code should be
reorganized to avoid having to use 'goto' at all.
-*- ReadFntFile() now ensures that the font metrics are valid before
trying to read the font file contents. In particular, it asserts that
the width, height, and bytes-per-char have valid values.
-*- Performed a number of very obvious optimizations.
-*- Simplified and cleaned up a lot of the code; but ReadFntFile() still
needs a reorganization (as noted earlier with regard to 'goto').
-*- Downgraded severity level of diagnostic messages from FATAL to WARNING.
127 lines of code changed in:
Made the following changes to csVector:
-*- No longer tries to malloc() zero bytes, as that will crash on some
platforms.
-*- No longer tries to realloc() a null pointer, as that will crash on
some platforms (rather than performing malloc as documented).
-*- Now checks for null pointer before calling free().
-*- Now enforces valid values for 'ilimit' and 'ithreshold'.
22 lines of code changed in:
Added scripts/python/*.py to installation.
I used Eric's improvements, so that the expensive operations are guarded.
24 lines of code changed in:
Update.
10 lines of code changed in:
Fix for particle systems in STAT_BSP worlds. Their bsp bounding boxes would
otherwise remain uninitialised and so always invisible. The loaders call to
Movable.UpdateMove () has no effect as the sectors are not yet prepared.
24 lines of code changed in:
Made compilable for Mac.
1 lines of code changed in:
Some cleanup.
1 lines of code changed in:
cspython plugin now also looks at the install path with
iSystem->GetInstallPath() to see where it should load the standard
CS Python support scripts (i.e. in scripts/python). So this means
that it is easier to use Python from external applications (games
outside of the CS directory structure).
12 lines of code changed in:
set void return type to SetPosition (gcc complains and vc sets int as return type implicitly (i think))
2 lines of code changed in:
added gamecore plugin to dynamic plugins (disabled since plugin does not compile yet)
1 lines of code changed in:
- Extended iPolygon3D interface with one of the SetTextureSpace
functions so that I can get the python script pysimp.py working.
- Added iMaterialWrapper interface so that there is now an SCF
interface for csMaterialWrapper.
- Added iWorld::FindMaterial() to look for a loaded material.
- Fixed iPolygon3D::GetVertex (and the W/C versions). They were
completely broken.
- Updated cs.i input file for SWIG so that it is more up-to-date
with regards to the current state in CS. The interface is far
from complete. First we need more and better defined SCF interfaces
for the engine itself and second we need to extend cs.i to actually
use all those interfaces. At this moment cs.i is extended
just enough so that I could get pysimp.py working.
- The pysimple test application now behaves almost exactly like
simple (without the sprite though) but the geometry of the
room is created in python instead of C++. Check out
scripts/python/pysimp.py to see how this works.
- SWIG generates a file called cs_wrap.doc. I also added this file
to CVS since it is useful to read and not everybody has SWIG.
5428 lines of code changed in:
Removed the unused, unfinished, and broken Macintosh Glide canvas. By all
accounts, it appears that this driver has never worked. Most of the
content of these source files is actually old code left over from X-windows
and Win32.
4 lines of code changed in:
Added a LoadFont(Name, File) method to G2D.
It returns the new fontid or -1 on failure.
Use this to load fonts.
30 lines of code changed in:
Norman Kramer updated config access in freefont plugin according
to Andrew's latest changes.
2 lines of code changed in:
Some experimentation with pysimple. Tried to update it so that
it works more like the normal simple but instead it creates
part of the geometry in python. However it does not work yet.
92 lines of code changed in:
- Made cspython plugin a real plugin (i.e. let it inherit from
iPlugIn).
- Fixed a bug in the cspython plugin which would initialize
System to NULL. I have no idea why that code was there.
19 lines of code changed in:
- Added a change from Luca Pancallo to 3ds2lev so that floating
point values near zero are better printed.
- Added '-l' option to 3ds2lev. If you use that option the
LIGHTING(no) will not be added.
11 lines of code changed in:
Minor fix - Shifted #include "cssysdef.h" back to
top of file, eliminating compile errors in VC++
2 lines of code changed in:
Fixed a bunch of compilation errors following Andrew's recent Big Change.
6 lines of code changed in:
Once again compiles following Andrew's Big Change(tm).
1 lines of code changed in:
Documented the 'make all' target in the "pseudo targets" section.
(How could we have possibly missed this one before? It's _so_ obvious. :-)
1 lines of code changed in:
- Fixed a bug in the 3dslev convertor where it would accidently
remove small polygons.
- Fixed two bugs in the new UV coordinate setter. The new code
was actually just ignored.
4 lines of code changed in:
Updated
7 lines of code changed in:
fixed radiosity when no polygons to light
3 lines of code changed in:
Fixed 'delete' versus DecRef() bugs.
4 lines of code changed in:
Fixed bug: Was leaking iConfigFile.
3 lines of code changed in:
Fixed bug: Was performing 'delete' on iConfigFile rather than DecRef().
4 lines of code changed in:
Eliminated a slew of compilation warnings in 3ds2lev.
7 lines of code changed in:
Fixed a crash bug in lighting if precalc_info was missing.
This bug was caused by Andrew's changes.
5 lines of code changed in:
- Andrew Zabolotny fixed vshell so that it compiles again after
Andrew's latest changes.
- Andrew also fixes system so that it is allowed to have
apps with no config files. This fixes the crash bug in simple.
23 lines of code changed in:
Some small cosmetic changes.
17 lines of code changed in:
Console fixes.
4 lines of code changed in:
Was still referencing obsolete crystalspace.console.stdout rather than
new crystalspace.console.output.standard.
1 lines of code changed in:
Fixed a potential memory leak in memfile.cpp which was introduced during
Andrew's Big Change(tm) today.
6 lines of code changed in:
fixes
2 lines of code changed in:
some fixes.
1 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