summaryrefslogtreecommitdiff
path: root/source4/pidl/lib
AgeCommit message (Collapse)AuthorFilesLines
2007-10-10r10829: Documentation updatesJelmer Vernooij6-26/+158
Update TODO Some small fixes to the modules (This used to be commit 0c53e7c3cf7fd91fd34c48a5e68c1bcf70569854)
2007-10-10r10788: - Give unions a nameJelmer Vernooij2-2/+3
- Check initialisation function return value (This used to be commit ebb8e75ebd572d62bdb3615d41c210bcc918fa41)
2007-10-10r10787: Fix silly nodiscriminant-issueJelmer Vernooij1-1/+1
(This used to be commit 43a5c863bf81af9b8415b6ccf8a386c36d4fa7f9)
2007-10-10r10765: Actually use VALS() setting for enumsJelmer Vernooij1-21/+21
(This used to be commit 6b33e3eedd13ade5c7c9f4b5062e8b8f919fe39c)
2007-10-10r10761: we need to use a pointer to a nbt_name to fix compiler warnings, ↵Stefan Metzmacher1-2/+2
because we can only use a pointers to unknown types in proto.h metze (This used to be commit 2f46e54e1bcf43f1bee062ff9a21e646cc3676e9)
2007-10-10r10760: remove only one star from the pointers, and not from the stars thatStefan Metzmacher1-1/+3
come from the array nesting this fixes a compiler warning and but in the eventlog.h, where the idl uses nstring strings[num_strings] metze (This used to be commit c4a252bd4678ef991705aaaf91c03c697f2f568e)
2007-10-10r10742: Support multi-level pointers + ref pointer fixesJelmer Vernooij4-25/+71
(This used to be commit 258b762dc62b257f99d1d859c5a3d850aba3e9fa)
2007-10-10r10739: Reduce number of calls to prs_align_custom()Jelmer Vernooij2-22/+26
(This used to be commit ebeeec5406308d493d45b1088963a87cdb953cac)
2007-10-10r10737: Fix some alignment issuesJelmer Vernooij2-43/+70
(This used to be commit 2b4270a2e965eb5c6fe0b25d22d2977e60bb7d43)
2007-10-10r10734: Generate ptr, size, offset, and length elements in unions just once.Jelmer Vernooij2-12/+30
(This used to be commit 12bfa5d01bcb4cb9dad5167e1a3721fd68f06275)
2007-10-10r10732: Generate _ctr structures for unionsJelmer Vernooij3-8/+65
(This used to be commit bd8fcb05003ad75f521783ad9603c923eacafc1a)
2007-10-10r10721: Handle allocations and primitive / deferred data correctly. In theory,Jelmer Vernooij3-83/+273
the generated output for DFS should work now (it compiles cleanly, but I haven't tested it yet). Not supported: - subcontexts() - relative pointers - unions of pointers - DATA_BLOB - several other things Also still need to do: - Remove some spurious spaces in the output - Do range() checking Example output is still available at http://samba.org/~jelmer/pidl_samba3/ (This used to be commit e2d7e382bb645f1bddd2047669bed10f121b59d2)
2007-10-10r10718: Another large set of small improvements. All generated files compileJelmer Vernooij4-56/+160
without warnings now. The only things left to do that are required for DFS: - add allocation of arrays in marshalling phase - handling primitive and deferred data in embedded structures / unions. Example output is again available from http://samba.org/~jelmer/pidl_samba3/ (This used to be commit 9fe724f6fb026d95306587f696c065f348aaf219)
2007-10-10r10717: Another bunch of small updates. All generated files except parse_dfs.cJelmer Vernooij5-19/+31
compile now when generated from Samba4's dfs.idl. (This used to be commit 2b315b1942e77640eca8e0a3347b2a9ed1920f6a)
2007-10-10r10716: Use correct Samba3 data types for strings. Also use Samba3 typesJelmer Vernooij1-22/+65
for a couple of other types (policy handles, SIDs, times) (This used to be commit c2527217b4b4c120d82044e65b979dd3b7d2609e)
2007-10-10r10715: More Samba3 parser generator improvements:Jelmer Vernooij2-30/+104
- Actually generate parsers for unions and structs. - Support some more builtin types. - Some more work on supporting arrays. - Several other small fixes. I've updated the example output at http://samba.org/~jelmer/ (This used to be commit b229c033ebc7ec972b32f1b75b60a9c68a36db97)
2007-10-10r10713: Couple more updates to the Samba3 parser generators.Jelmer Vernooij5-90/+300
Unions and enums have been improved, init functions are now generated properly, some other small improvements. (This used to be commit 8a60e79175eb27ef9fa4b8dea72a518bbaab900f)
2007-10-10r10698: Fix support of enums in switch_type() in the ethereal parser generatorJelmer Vernooij1-4/+13
(This used to be commit 9fac8dfeb393212e16be7156de466077bc90232d)
2007-10-10r10694: Add some work I did this afternoon on getting pidl to output Samba3Jelmer Vernooij9-40/+780
RPC parsers. Currently the following files can be generated: - include/rpc_BASENAME.h - rpc_server/srv_BASENAME.c - rpc_server/srv_BASENAME_nt.c (template only, user has to fill in functions) - rpc_client/cli_BASENAME.c - rpc_parse/parse_BASENAME.c So far, I have been working on getting DFS working. Currently still to do (all in rpc_parse/parse_BASENAME.c): - Proper handling of declarations - Proper handling of scalar/buffer parts of structs and unions - Subcontexts - Proper handling of arrays - Support for custom (non-scalar) types I hope to have a somewhat more working version later this week. Some files as currently generated are available from: http://samba.org/~jelmer/pidl_samba3/ (This used to be commit 8af8eaeeef6d46f4d25ccb1d25890e1eef063e4f)
2007-10-10r10608: - fix hierachical memory handling in ndr_pull_nbt_nameStefan Metzmacher1-3/+8
- add wrepl_nbt_name scalar type and do the pull/push in the ndr layer instead of the caller - give the flags and group_flag in the wrepl_name a meaning metze (This used to be commit b98efc2905e1147eb97111b46a877bdb9d8dd154)
2007-10-10r10388: Add version numbers (required for CPAN)Jelmer Vernooij22-0/+60
(This used to be commit 786329576bf14ae774f9d5a24268e46b7dcb634a)
2007-10-10r10296: Fix function pointer handling for older perl versionsJelmer Vernooij1-1/+1
(This used to be commit 0949db8119af548c199930c64449d6ca2228bcb5)
2007-10-10r10295: Remove dependency on Data::DumperJelmer Vernooij1-1/+0
(This used to be commit 26f2903dab833fbf420af7104757b26f8398e99d)
2007-10-10r10197: Assume that external dissectors are structs which I think is always theTim Potter1-1/+1
case. (This used to be commit 515f456facfeb2cd9893ac65dd96cdda16c9d1ca)
2007-10-10r10123: Add more warnings. Support quotes in conformance command argumentsJelmer Vernooij3-26/+60
(This used to be commit e6842fcc9809bcf8de678199a6f28fbbde6c0b83)
2007-10-10r10116: Print out a couple more warnings.Jelmer Vernooij2-4/+57
(This used to be commit 0628dfa5c3c74614a86b4f61b8d1555ef41c41bb)
2007-10-10r10110: Add some more warnings, implement FIELD_DESCRIPTIONJelmer Vernooij2-21/+91
(This used to be commit 1caeb3238dac6321bde8e254a8efaf090b4d56b0)
2007-10-10r10094: Support quoted arguments in conformance filesJelmer Vernooij1-2/+3
Update pidl manpage (This used to be commit a69e88e7b19b8f05222b54aea88395b51b96c003)
2007-10-10r10093: Fix the HF_FIELD conformance file commandJelmer Vernooij3-15/+15
(This used to be commit 0c0a4b55cff4079276073060dae91ff0c19af42f)
2007-10-10r10074: Pass CPP to pidl via environment variables rather then config.pm. ↵Jelmer Vernooij1-2/+1
This fixes the standalone pidl build (as used for ethereal) (This used to be commit 9c9ebd2214423c58357854f09bd744e13e807d8f)
2007-10-10r10071: Configure checks for IRIX build environment. Test whether we canJames Peach1-1/+2
use the MIPSPro 7.4 -c99 option to get C99 support. Try to find a common perl that is more modern than /usr/bin/perl. (This used to be commit 82fab8b747bf68d8548c6f0f2f4bff98bd428d22)
2007-10-10r10047: Translate an idl NTTIME to an ethereal FT_ABSOLUTE_TIME instead ofTim Potter1-2/+6
a FT_UINT64. Not sure about a NTTIME_hyper though. (This used to be commit 42568d3f286679656417301d1cf29d3f0cd71030)
2007-10-10r10030: Add hierarchical memory allocation to TDR's pull functionsJelmer Vernooij1-8/+15
(This used to be commit 0e358de93a007db921ad5e9a892114122254de39)
2007-10-10r10015: Change the NT4 registry backend to use the IDL-generate parse functions.Jelmer Vernooij1-602/+785
(This used to be commit 0ad46ef804c0654e927f9c14ea93c45f9e3c718c)
2007-10-10r10010: Support base_type() attributeJelmer Vernooij1-1/+3
(This used to be commit c0bae9b0b5cb56aea98e33b4d0a9d29c6622dd27)
2007-10-10r10006: Insert correct header file.Jelmer Vernooij1-5/+8
(This used to be commit dfa45ec1b329ad01daf81beeb31985811ce4f5af)
2007-10-10r9984: Honour the NOEMIT conformance directive for entire IDL files.Tim Potter1-0/+2
(This used to be commit 73dffade2c008b26eaf3924d31693c27328b8586)
2007-10-10r9983: Make the NOEMIT conformance directive with no argument mean do notTim Potter1-2/+9
emit any dissectors for this IDL. (This used to be commit 55a98b3dfdef21df172dac6cff4ebc114b35df00)
2007-10-10r9982: Strip prefixes off the check for noemit in the conformance info so youTim Potter1-4/+6
can use the typedef names in the IDL rather than the autogenerated function names. This means you can say "NOEMIT security_descriptor" instead of "NOEMIT dissect_security_descriptor_type" or whatever. (This used to be commit ec80c5ceec3c96dc847d171967472cce99edcdc7)
2007-10-10r9734: The ethereal parser generator was generating duplicate functions forTim Potter1-4/+4
fields in structures with the same name as a structure (i.e security_ace.object and security_ace_object). I've twiddled the naming scheme a bit and things are a bit more unique but there is still may be some naming conflicts in other IDL files. We are now getting confused over fields with the same name in unions (e.g security_ace_object_ctr.object) plus some other union weirdness. (This used to be commit c2ad9f49a204d0075a3b460c5ea2ca9741fc7125)
2007-10-10r9683: OK that last fix was too good to be true. (-:Tim Potter1-1/+1
(This used to be commit b33774b693f9866295fc11758c826ff150bfb02d)
2007-10-10r9682: Jelmer, I think this fixes another case where the parser generator wasTim Potter1-1/+1
producing duplicate function definitions. Can you check this out? (This used to be commit e9bb695e74c98ea1f2647dacc7c9c4b8ad37bd72)
2007-10-10r9677: Don't generate duplicate function names when processing structure ↵Tim Potter1-3/+3
elements that are bitmaps or enums. (This used to be commit 1b0e65359cc9caee56a63690af142e8599ea6926)
2007-10-10r9612: Don't generate handoff code for idl files that don't implement a pipeTim Potter1-8/+11
(e.g security.idl) (This used to be commit 178d040dbb07f53c29e945a2109bb3c8f355e3d3)
2007-10-10r9611: Reset global variables (hf and ett lists) for each parser run. ThisTim Potter1-6/+4
avoids building up buckets of unecessary definitions from other parsers. Get rid of warning and FIXME for unknown data types. This is handled by the interface depends list and including the appropriate header files. (This used to be commit eb5e3b031cd18fb1f83d9ccbb21e5148c8c1cdd0)
2007-10-10r9576: Couple of small fixes. Generate notice indicatingJelmer Vernooij1-5/+18
that the parsers were autogenerated. (This used to be commit a37e2134e61eb38e9cbc54f8533113622f013037)
2007-10-10r9558: Add more whitespace to generated output.Tim Potter1-3/+3
Change conformance file warning to something more understandable. Don't generate duplicate duplicate element dissectors for function call arguments. Hey this makes the winreg dissector compile, but not link. (This used to be commit 545d3deaa43ed2e837a7fb5e0cf5761281a9451f)
2007-10-10r9554: Do a quick once-over to make the ethereal auto-generated parser code aTim Potter1-51/+54
bit more readable. (This used to be commit da3ca9e3c1df09983629eeee8baab2196f2135a9)
2007-10-10r9553: Some cosmetic things to make the atsvc dissector output look prettierTim Potter1-6/+18
and be more usable: - nicer formatting for summary of set fields in bitmap. - say "Pointer to Foo" in proto_tree instead of just "Foo" when dissecting pointers. - append string value to proto_tree when dissecting pointers to strings. - strip librpc/gen_ndr prefix from includes so generated files can live in the ethereal epan/dissectors directory. Now to work on getting the winreg dissector compiling... (This used to be commit ab00976699a2deeb3ddca510864520fcb46714bc)
2007-10-10r9519: Use the value() value of an element when that element is used inJelmer Vernooij1-0/+21
limited expressions (size_is,length_is,subcontext_size,etc) (This used to be commit 886780c298a794f304b0fce851bbb58c53605d17)