summaryrefslogtreecommitdiff
path: root/doc/map_cache.txt
diff options
context:
space:
mode:
authorDracoRPG <DracoRPG@54d463be-8e91-2dee-dedb-b68131a5f0ec>2007-04-06 17:22:40 +0000
committerDracoRPG <DracoRPG@54d463be-8e91-2dee-dedb-b68131a5f0ec>2007-04-06 17:22:40 +0000
commitded1f674755335a9cbb5c08e321b4a4c092a0e8d (patch)
tree4ff5231731935b41f9ef1fdccf9b14ef5b4e85b3 /doc/map_cache.txt
parent2d75543b172d4cf141278bbd224e2353f19b5c86 (diff)
downloadhercules-ded1f674755335a9cbb5c08e321b4a4c092a0e8d.tar.gz
hercules-ded1f674755335a9cbb5c08e321b4a4c092a0e8d.tar.bz2
hercules-ded1f674755335a9cbb5c08e321b4a4c092a0e8d.tar.xz
hercules-ded1f674755335a9cbb5c08e321b4a4c092a0e8d.zip
I'm still here!
Rewrote fame ranking lists- changed MAP_NAME_LENGTH to 12, now there's MAP_NAME_LENGTH_EXT at 16 for uses where there is / may be the .gat extension, code adjusted accordingly - moved map_normalize_name to mapindex_normalize_name so that everything handling map names uses the same extension-removing function - greatly enhanced the map cache generator, complete documentation on the tool and the map cache format can be found in doc/ - the map cache format changed a bit as a consequence, but of course a new valid one is included (contains latest Nameless Island maps) - fixed a duplicate entry in map index git-svn-id: https://rathena.svn.sourceforge.net/svnroot/rathena/trunk@10167 54d463be-8e91-2dee-dedb-b68131a5f0ec
Diffstat (limited to 'doc/map_cache.txt')
-rw-r--r--doc/map_cache.txt71
1 files changed, 60 insertions, 11 deletions
diff --git a/doc/map_cache.txt b/doc/map_cache.txt
index 4ab77e99b..82849cdf4 100644
--- a/doc/map_cache.txt
+++ b/doc/map_cache.txt
@@ -1,18 +1,67 @@
-"How to use the mapcache builder"
-DracoRPG
+//===== Athena Doc ========================================
+//= Map Cache Builder and Format Documentation
+//===== By ================================================
+//= DracoRPG
+//===== Version ===========================================
+//= 1.0
+//=========================================================
+//= 0.1 - Short howto for the initial builder version
+//= 1.0 - Complete manual covering the improved version
+//===== Description =======================================
+//= A complete manual for eAthena's map cache generator
+//= as well as a reference on the map cache format used
+//=========================================================
+Preface:
+-------------------------------------------------------------------------------
-This is only useful if you have custom maps, as eAthena is provided with an updated mapcache containing every map
-from kRO Sakray's data.
+Since SVN revision ~10000, the map-server does not know how to read RO client files anymore. It reads maps from a
+"map cache" file that contains all and only the useful data about the maps. A map cache containing every official
+kRO Sakray map currently supported by eAthena is provided as a default.
+If you have custom maps or want to minimize the size of your map cache because your server does not load all of them
+(multi-map-server or light test server), you can use the map cache builder to generate a new one fitting your needs.
-1. First add the path to the directory / GRF containing your maps to db/grf_files.txt
-/!\ Please note you must also have the official maps as the whole mapcache will be rebuilt from scratch
+Map cache builder manual:
+-------------------------------------------------------------------------------
-2. Then add those custom maps at the end of db/map_list.txt, carefully chosing their index
+The source code for the map cache builder is located in src/tool/. It can be built using "make tools" if you use the Makefile
+or using the "mapcache" project under Visual Studio. Named "mapcache", the executable will be in your eAthena main folder.
+The map cache builder needs 3 file paths : one is a list of GRFs and/or data directory containing the maps, the second
+is the list of maps to add to the map cache, and the last one is the path of the map cache to generate. Default values for
+those paths are "tools/mapcache/grf_files.txt", "db/map_index.txt" and "db/map_cache.dat".
+The list of GRF and/or data directory must follow the format and indication of the default file: as many "grf:" entries as
+you wish and optionally one only "data_dir:" entry with trailing backslash included. // comments are supported as usual.
+In fact, any file with one map name per line can be used as a map list, that's why the map index list is used as a default:
+we are sure it contains every map supported by the server. Anything after the map name is ignored, // comments are supported
+and if the first word on the line is "map:" then the second word is used as the map name instead: that allows using
+maps_athena.conf as your map list, which is handy if you want to generate a minimal map cache for each of your multiple
+map-servers.
+The map cache file path can point to an already existing file, as the builder adds a map only if it's not already cached.
+This way, you can add custom maps to the base map cache without even needing kRO Sakray maps. If you wish to rebuild the
+entire map cache, though, you can either provide a path to a non-existing file, or force the rebuild mode.
-3. Now just run the mapcache builder and it'll build a new one at db/map_cache.dat
+Here are the command-line arguments you can provide to the map cache builder to customize its behavior:
+ -grf path/to/grf/list
+ Allows to specify the file containing the list of GRFs and/or data directory
+ -list path/to/map/list
+ Allows to specify the file containing the list of maps to add to the map cache
+ -cache path/to/map/cache
+ Allows to specify the path to the generated map cache
+ - rebuild
+ Allows to force the rebuild mode (map cache will be overwritten even if it already exists)
-NOTE:
-You can override those default paths by providing your own ones as command-line arguments to the mapcache builder:
-$> mapcache [grf_files_path [map_list_path [map_cache_path]]] \ No newline at end of file
+Map cache format reference:
+-------------------------------------------------------------------------------
+
+The file is written as little-endian, even on big-endian systems, for cross-compatibility reasons. Appropriate conversions
+are done when generating it, so don't worry about it.
+The first 6 bytes are a main header:
+<unsigned long> file size
+<unsigned short> number of maps
+Then maps are stored one right after another:
+<12-characters-long string> map name
+<short> X size
+<short> Y size
+<long> compressed cell data length
+<variable> compressed cell data