Is this trivial patch OK?
- Julian
* subversion/libsvn_wc/wc_db.h
Document the conventions about *_ABSPATH parameters at the top of the file.
Remove obsolete "###" comments about this and about pool conventions.
Index: subversion/libsvn_wc/wc_db.h
===================================================================
--- subversion/libsvn_wc/wc_db.h (revision 904098)
+++ subversion/libsvn_wc/wc_db.h (working copy)
@@ -85,6 +85,10 @@
function, and maintains information and state about every working
copy "touched" by any of the APIs in this interface.
+ * *_ABSPATH
+ All *_ABSPATH parameters in this API are absolute paths in the local
+ filesystem, represented in Subversion internal canonical form.
+
* LOCAL_ABSPATH
This parameter specifies a particular *versioned* node in the local
filesystem. From this node, a working copy root is implied, and will
@@ -257,12 +261,6 @@
} svn_wc__db_lock_t;
-/* ### note conventions of "result_pool" for the pool where return results
- ### are allocated, and "scratch_pool" for the pool that is used for
- ### intermediate allocations (and which can be safely cleared upon
- ### return from the function).
-*/
-
/* ### NOTE: I have not provided docstrings for most of this file at this
### point in time. The shape and extent of this API is still in massive
### flux. I'm iterating in public, but do not want to doc until it feels
@@ -271,16 +269,6 @@
/* ### where/how to handle: text_time, locks, working_size */
-/* ### update docstrings: all paths should be internal/canonical */
-
-/**
- * ### KFF: Would be good to state, here or in an introductory comment
- * ### at the top of this file, whether subsequent 'path' parameters
- * ### are absolute, or relative to the root at which @a *db was
- * ### opened, or perhaps that both are acceptable.
- */
-/* ### all paths in this API are absolute, and internal/canonical form. */
-
/**
* @defgroup svn_wc__db_admin General administractive functions
