Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | make some more things private | Jesse Luehrs | 2013-02-28 | 2 | -14/+13 | |
| | ||||||
* | try to name this a bit better | Jesse Luehrs | 2013-02-27 | 6 | -18/+22 | |
| | ||||||
* | add a leading newline to diags when running under a non-raw reporter | Jesse Luehrs | 2013-02-27 | 5 | -33/+70 | |
| | | | | | makes things easier to read this way, when diags are printed to stderr interspersed with test summary information | |||||
* | use the new stream parsing stuff in the summary reporter | Jesse Luehrs | 2013-02-27 | 1 | -16/+66 | |
| | ||||||
* | this name parameter was never used | Jesse Luehrs | 2013-02-27 | 2 | -23/+8 | |
| | ||||||
* | short summary goes after long summary | Jesse Luehrs | 2013-02-27 | 1 | -1/+1 | |
| | ||||||
* | use iterators instead of streams | Jesse Luehrs | 2013-02-27 | 1 | -16/+16 | |
| | | | | | | | | | streams buffer the next character (to support Stream.Empty), which means that if there is exactly one line available, it will read that line and then block waiting for the next character before the result stream can be returned. this means that we don't get results from the current line until the next line is ready. iterators don't have that problem, since they don't have a separate Empty type, they just have a hasNext method. | |||||
* | also generate start and end events | Jesse Luehrs | 2013-02-27 | 1 | -2/+9 | |
| | ||||||
* | generate plan and result events | Jesse Luehrs | 2013-02-26 | 1 | -3/+9 | |
| | | | | ignore comments and subtests completely for now | |||||
* | these don't need to be public | Jesse Luehrs | 2013-02-26 | 1 | -2/+2 | |
| | ||||||
* | use a recursive call rather than a var | Jesse Luehrs | 2013-02-26 | 1 | -18/+23 | |
| | | | | | a bit more complicated than i'd like, due to path-dependent types, but oh well | |||||
* | more cleanup | Jesse Luehrs | 2013-02-26 | 1 | -13/+7 | |
| | ||||||
* | split out the tap stream parser from the individual line parsers | Jesse Luehrs | 2013-02-26 | 4 | -299/+283 | |
| | ||||||
* | make this a lazy val too | Jesse Luehrs | 2013-02-26 | 1 | -1/+1 | |
| | ||||||
* | more refactoring | Jesse Luehrs | 2013-02-26 | 1 | -50/+52 | |
| | ||||||
* | simplify | Jesse Luehrs | 2013-02-26 | 1 | -94/+53 | |
| | ||||||
* | make a couple more traits sealed | Jesse Luehrs | 2013-02-26 | 2 | -3/+3 | |
| | ||||||
* | simplify | Jesse Luehrs | 2013-02-26 | 1 | -5/+6 | |
| | ||||||
* | using streams is cleaner here | Jesse Luehrs | 2013-02-26 | 1 | -37/+36 | |
| | ||||||
* | rewrite the tap parser | Jesse Luehrs | 2013-02-26 | 1 | -77/+313 | |
| | | | | | | | | this parser is line based, which should make the top level logic easier to understand (although it does push some hairy stuff down to the bottom that i still need to clean up). it also should be able to be easily modified to work as a streaming parser, although i haven't tried to do that yet. | |||||
* | make the test method hiding stuff a bit less fragile. hopefully. | Jesse Luehrs | 2013-02-25 | 1 | -23/+36 | |
| | ||||||
* | make this a bit cleaner | Jesse Luehrs | 2013-02-25 | 1 | -1/+1 | |
| | ||||||
* | style | Jesse Luehrs | 2013-02-25 | 4 | -4/+4 | |
| | ||||||
* | comment that, because it's a bit non-obvious | Jesse Luehrs | 2013-02-25 | 1 | -0/+9 | |
| | ||||||
* | make sure stdout and stderr are synchronized when possible | Jesse Luehrs | 2013-02-25 | 1 | -1/+7 | |
| | ||||||
* | switch back to (a simpler form of) tests reporting their exit code | Jesse Luehrs | 2013-02-25 | 5 | -18/+18 | |
| | ||||||
* | allow running external programs as tap producers | Jesse Luehrs | 2013-02-24 | 1 | -0/+41 | |
| | ||||||
* | a bit more cleanup | Jesse Luehrs | 2013-02-23 | 4 | -5/+5 | |
| | ||||||
* | clean up imports | Jesse Luehrs | 2013-02-23 | 11 | -73/+55 | |
| | ||||||
* | simplify | Jesse Luehrs | 2013-02-23 | 1 | -3/+2 | |
| | ||||||
* | clean up plan object creation a bit | Jesse Luehrs | 2013-02-23 | 3 | -5/+5 | |
| | ||||||
* | tests shouldn't produce anything other than tap | Jesse Luehrs | 2013-02-23 | 8 | -47/+72 | |
| | | | | the harness can parse the tap to figure out whatever it needs from that | |||||
* | start working on docs | Jesse Luehrs | 2013-02-23 | 1 | -0/+5 | |
| | ||||||
* | allow extensions to get the correct stack trace | Jesse Luehrs | 2013-02-23 | 1 | -3/+24 | |
| | | | | better implementation ideas here would be nice | |||||
* | don't need these | Jesse Luehrs | 2013-02-23 | 1 | -2/+0 | |
| | ||||||
* | diag should print to stderr, so it's visible (note is for stdout) | Jesse Luehrs | 2013-02-23 | 1 | -10/+19 | |
| | ||||||
* | clean up the test runner | Jesse Luehrs | 2013-02-23 | 2 | -26/+75 | |
| | ||||||
* | stop threading outputstreams through everything | Jesse Luehrs | 2013-02-23 | 5 | -35/+19 | |
| | | | | | it's not really necessary, since we can already capture output if necessary externally | |||||
* | make this pass the outputstream for now | Jesse Luehrs | 2013-02-23 | 1 | -1/+3 | |
| | | | | will clean this up later | |||||
* | oops, this was broken | Jesse Luehrs | 2013-02-23 | 1 | -2/+2 | |
| | ||||||
* | integrate with sbt | Jesse Luehrs | 2013-02-23 | 3 | -0/+64 | |
| | ||||||
* | a bunch of refactoring and cleanup | Jesse Luehrs | 2013-02-22 | 1 | -27/+78 | |
| | ||||||
* | give this a bit more space | Jesse Luehrs | 2013-02-22 | 1 | -1/+1 | |
| | ||||||
* | hack together a multi-test harness | Jesse Luehrs | 2013-02-22 | 5 | -24/+146 | |
| | | | | needs a *lot* of work, but the basic idea is there | |||||
* | tap parser | Jesse Luehrs | 2013-02-22 | 1 | -0/+142 | |
| | ||||||
* | start fleshing out the harness | Jesse Luehrs | 2013-02-21 | 5 | -4/+59 | |
| | ||||||
* | move more things around | Jesse Luehrs | 2013-02-21 | 4 | -12/+15 | |
| | ||||||
* | handle mismatches plans | Jesse Luehrs | 2013-02-20 | 1 | -3/+20 | |
| | ||||||
* | cleanups | Jesse Luehrs | 2013-02-20 | 1 | -5/+4 | |
| | ||||||
* | oops | Jesse Luehrs | 2013-02-20 | 1 | -1/+1 | |
| |