summaryrefslogtreecommitdiff
path: root/lib
Commit message (Collapse)AuthorAge
* fix some conditional-compilation issues that trip -Wunreachable-codeGravatar Karl Ramm2009-04-30
|
* Set up a local UDP socket to drive testingGravatar Karl Ramm2009-04-27
| | | | | | | | | | | | | Using socketpair doesn't work because you end up with an AF_UNIX socket, and it even if you _can_ use sendto with anonymous unix-domain sockets (which the documentation is mute on and it doesn't seem worth experimenting on), the zephyr library doesn't allocate enough storage for a sockaddr_un, and although I'm not above fixing that, I determined that for structural reasons it's too annoying right now. (But once we have a test suite...) Anyway, cons up a udp socket, and extract a port number from the library, and use the python socket interface to send the zephyr packet.
* It turns out that sockpair /does/ work with AF_UNIX, SOCK_DGRAM, just notGravatar Karl Ramm2009-04-27
| | | | | | | | AF_INET, which is not terribly surprising now that I think about it. However, ZPending() still returns -1, and I don't know how to debug why at the moment.
* draft packet tests - using pipe trick turns out not to work, though.Gravatar Mark W. Eichin2009-04-26
|
* take ZGetDestAddr off the listGravatar Mark W. Eichin2009-04-24
|
* test that the file descriptor ZGetFD returns is actually a socketGravatar Mark W. Eichin2009-04-24
|
* test ZGetDestAddrGravatar Mark W. Eichin2009-04-24
|
* use less oddball typesGravatar Karl Ramm2009-04-21
|
* populate_enum - instantiate the enum values as actual objectsGravatar Mark W. Eichin2009-04-18
| | | | | | (in 2.6, do this with a class decorator?) expand the ZFormatNotice/ZParseNotice round trip, but it needs more...
* More i18n infrastructureGravatar Karl Ramm2009-04-17
| | | | | | | | | | | | | | | | lib/charset.c h/zephyr/zephyr.h Refactor ZGetCharset into ZGetCharset and a function that does all the same defaulting but returns a string. ZTransliterate is a zephyr-metaphor wrapper for iconv. lib/zephyr-tests.py Wrappers for the functions in charset.c (but not tests) lib/zephyr_tests.txt Tests for the functions in charset.c
* basic Zcode testGravatar Mark W. Eichin2009-04-17
|
* more api todosGravatar Mark W. Eichin2009-04-17
| | | | | clarify which ZLocateUser actually exists
* show what ZMakeAscii actually *does*Gravatar Mark W. Eichin2009-04-16
|
* wrap ZMakeAsciiGravatar Mark W. Eichin2009-04-16
| | | | | | | | | clone it too test the wrapper against the clone fix typo in description add todos add future-coverage list
* Clean up $Zephyr$sGravatar Karl Ramm2009-04-14
|
* Export a new variable, Zauthtype, which indicates what sort of authenticationGravatar Karl Ramm2009-04-14
| | | | | | | | | the library was compiled for. Use this variable in the tests such that they will pass (and, like, test something) if the library was not compiled with authentication.
* should not be exportedGravatar Karl Ramm2009-04-14
|
* test ZOpenPort and ZClosePortGravatar Mark W. Eichin2009-04-14
| | | | | record coverage, start a checklist of what next
* ZOpenPort, ZClosePortGravatar Mark W. Eichin2009-04-14
|
* backgroundGravatar Mark W. Eichin2009-04-14
|
* allow zephyr_run_doctests -v to workGravatar Mark W. Eichin2009-04-14
|
* run the tests from makeGravatar Karl Ramm2009-04-14
|
* add defs for datarootdir so configure will Shut UpGravatar Karl Ramm2009-04-13
| | | | | | | | | | Rearrange what libraries get pulled in where, so dpkg-shlibdeps doesn't whine (see a theme here?) (Also so that, say, znol, doesn't have to link with Everything.) This will break platforms that don't do shared-library dependencies.
* draft doctest version of the existing tests, to show what they lookGravatar Mark W. Eichin2009-04-13
| | | | | | | | | | like; might serve as a start on a "live API guide" once it is fleshed out more. Emphasizes the commentary which actually makes a lot of sense (a bunch of things went into the svn commit logs which really should have been in-line in the test cases; this approach avoids that...)
* test ZExpandRealmGravatar Mark W. Eichin2009-04-10
|
* kstuff.c:Gravatar Karl Ramm2009-04-10
| | | | | | | | | Turn off the replay cache (at least with MIT krb5 1.6) All three authentication checks: Initialize cksum1_buf so gcc will shut up.
* m-x untabifyGravatar Mark W. Eichin2009-04-09
|
* move run from ZephyrTestSuite to TestSuite where it belongs (easier toGravatar Mark W. Eichin2009-04-09
| | | | | | | | | | replace that way too) add starting/done/failed to TestSuite.run output wrap ZCompareUID test ZCompareUID
* discovered by looking at code that uniqueid timestamps are in networkGravatar Mark W. Eichin2009-04-08
| | | | | | | | byte order. They're really bytestrings so it shouldn't matter... but since this is half testing, half debugging, and half API exploration, we might as well make them human-displayable...
* reworked with a pprint/pformat distinction (mostly to stop propating indent)Gravatar Mark W. Eichin2009-04-07
| | | | | | | added comments z_sender_sockaddr now displays only the branch implied by sa_family
* allow a class to override the display of individual fields; since thisGravatar Mark W. Eichin2009-04-07
| | | | | | | | | | | is at the class level, it can look at other fields (ie. z_num_hdr_fields) for lengths or other parameters that modify the display. implemented for z_hdr_fields, z_other_fields. probably useful for selecting z_sender_sockaddr values too.
* ctypes docs say "enums are easy" but don't give any further guidance.Gravatar Mark W. Eichin2009-04-07
| | | | | | | | Turns out they are easy :-) Also added "fake" enums to render address family tags correctly; cheat a little and lift the values from the python socket module directly, to avoid duplication.
* add type-specific pretty-printers (yay classes)Gravatar Mark W. Eichin2009-04-07
| | | | | | | | the mess around _ZTimeval.pprint is that either they're not getting initialized, not getting converted properly, or aren't really timestamps anymore.
* generic pretty-printer for ctypes Structures and UnionsGravatar Mark W. Eichin2009-04-07
| | | | | | use it to display fields of interest for testing fix the immediately exposed bug (tv_sec/tv_usec sign)
* allow callers to specify the installed shared libGravatar Mark W. Eichin2009-04-07
|
* split libzephyr wrappers apart from the tests; eventually they'llGravatar Mark W. Eichin2009-04-07
| | | | | | be a useful module on their own.
* fix keywords magic (by changing Id to Id: which won't be visible in the diff)Gravatar Mark W. Eichin2009-04-06
|
* unit tests for libzephyr.Gravatar Mark W. Eichin2009-04-06
| | | | | | | uses python ctypes to build structs and call functions; minimal current case is an empty ZFormatNotice/ZParseNotice round trip.
* be more paranoid about what fields we are checksummingGravatar Karl Ramm2009-04-04
|
* another phantom nullGravatar Karl Ramm2009-04-03
|
* apple error table initialization for #41Gravatar Karl Ramm2009-04-02
|
* Explicitly #include <arpa/nameser_compat.h> if it exists.Gravatar Karl Ramm2009-04-02
|
* cp zephyr_err.h out of the build directory rather than mv, because theGravatar Karl Ramm2009-04-02
| | | | | | MacOS X compile_et produces a zephyr_err.c that (mysteriously) includes "zephyr_err.h", so it has to stay in the directory.
* keep track of total # of incoming header fields, and cksum appropriatelyGravatar Karl Ramm2009-03-24
|
* Brain dumps used to stash the address of the client for which metadataGravatar Karl Ramm2009-03-23
| | | | | | | | | | | | was being sent in an extra header field. This worked about as well as you might expect. Fortunately we happened to be putting an address in header field 18 anyway, so backwards compatibility should be relatively painless. I don't know which approach is a more egregious hack, but this change at least nets us four less lines of code, which is obviously good, right? Oh, and while we're at it, if the sa_family isn't set on the sender address, just cringe and set it to AF_INET.
* notice->z_other_fields[notice->z_num_other_fields] is either going toGravatar Karl Ramm2009-03-23
| | | | | | | be NULL, something Wrong^TM, or off the end of z_other_fields. bad all around.
* return UNKNOWN if we don't recognize the charset #Gravatar Karl Ramm2009-03-22
|
* only run setlocale onceGravatar Karl Ramm2009-03-22
|
* more cmu stuffGravatar Karl Ramm2009-03-16
|
* file for charset stuffGravatar Karl Ramm2009-03-16
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-29 12:15:21 +0000