more documentation and cleanups for various modules.

icc/README

@@ -25,7 +25,7 @@ For information on ICC profiles, you may want to visit:
 
 == Notes for porters ==
 
-Ehe endianess and other platform specific configuration
+The endianess and other platform specific configuration
 are in:
 
 icc/<build>/misc/build/SampleICC-1.3.2/IccProfLib/IccProfLibConf.h
@@ -33,5 +33,3 @@ icc/<build>/misc/build/SampleICC-1.3.2/IccProfLib/IccProfLibConf.h
 The patchs:
 SampleICC-1.3.2.patch tweaks the build to use the endianness
 detected in sal.
-
-beppec56@openoffice.org

instsetoo_native/README

+native install-set creation
+
+This is where you will find your natively packaged builds after the
+build has completed. On windows these would live in:
+
+instsetoo_native/wntmsci12.pro/LibreOffice_Dev/native/install/en-US/*.msi
+
+for example (nothing like a few long directory names before breakfast).
\ No newline at end of file

io/README

-Simple IO wrappers.
+Simple IO wrapper UNO components
+
 

ios/README

+iOS specific code and tests

jpeg/README

-Support for jpeg-format. (Which library, where used ???)
+External library for reading/writing jpegs
+
+This is only used by the svtools jpeg import filter that is provided
+for use by VCL see [[svtools/source/filter/jpeg]]
+
+

languagetool/README

 Style and grammar checker for various languages written in Java, from [http://www.languagetool.org/]
+
+Shipping this is a compile-time option for LibreOffice, but our
+downloadable builds use the light-weight lightproof grammer checker
+instead to avoid the Java bootstrapping performance issues.
\ No newline at end of file

libexttextcat/README

+N-Gram-Based Text Categorization library
+
+This guesses the language that input text is written in. Fundamentally
+an adaption of of wiseguys libtextcat extended to be UTF-8 aware.
+
+It is hosted at: [http://cgit.freedesktop.org/libreoffice/libexttextcat/]

mdds/README

 Multi-dimensional data structure (mdds) library, available from [http://code.google.com/p/multidimalgorithm/].
+
+mdds primarily provides data structures that are used by the calc
+core.
\ No newline at end of file

packimages/README

+Packs used images in sorted order into theme .zip files
+
+The essence of this module is contained within [[solenv/bin/packimages.pl]].
+This script takes several sources of information:
+
+* the list of images used in every .src file that is compiled
+* the configuration XML describing every command that is used
+* [[packimages/pack/iamge-sort.lst]] which provides a profile based
+  ordering of images.
+
+It then tries to pack the images together into the .zip file into an
+order that matches / is similar to that which we want at run-time, ie.
+an entire toolbar's images should (ideally) be next to each other in a
+nice linear run, to reduce LibreOffice's working set, and improve
+startup performance: we require all these images to get paged in on
+start.

postprocess/README

 Postprocessing and checking of files delivered by other modules.
 
-This module hast to be the last one before creating install sets in module 'instset_native'.
+This module has to be the last one built before creating install sets
+in module 'instset_native'. Thus it ties together all the dependencies
+of all the other de-coupled modules. See the first line of
+[[postprocess/prj/build.lst]] for that.

psprint_config/README

-Contains ppds for use by psprint when not using CUPS.
+Contains ppds for use by vcl when not using CUPS
 
-(CH: psprint seems to have been gone ???)
+This is used for the print-to-file functionality. These two PPDs
+describe the range of paper sizes and postscript options necessary for
+printing to postscript without a configured printer.

registry/README

 Registry reading, etc.
+
+This provides tools for dealing with the legacy binary types database
+format, still in use by extensions and the core code. While the actual
+binary file format is implemented by the [[store]] code, the wrapper
+that turns this into a type registry is implemented here.
+
+While this code is primarily used in only two modes:
+
+* linear write / concatenate
+* random access read
+
+The API unfortunately exposes a random-access write approach, which -
+while ~unused needs before we can re-write this away from the store
+backend.

xmlscript/README

 XML dialogs.
+
+This code is used to (de)serialize basic dialogs to XML for storage
+inside documents. While the XML -appears- to have some hierarchical
+structure, that is only a fabrication, parsing and underlying toolkit
+widget structure is sadly linear and flat.
\ No newline at end of file

xmlsecurity/README

 Stuff for document signing.
+
+This code provides dialogs, and infrastructure wrapping libxmlsec that
+implements document signing.