Bring up-to-date w/ current Midje, cljs#7

Open

charles-dyfis-net wants to merge 240 commits intobaznex:masterfrom charles-dyfis-net:master

+3,185-3,226

Commits on Feb 19, 2012

[Issue marick#82 ] first version of formula: only reports 0-1 times per formula, regardless of # of generated fact runs
Alex Baranosky
committed
[Issue marick#82 ] removed unused code
Alex Baranosky
committed
[Issue marick#82 ] adding backlog file for the formula epic
Alex Baranosky
committed
[Issue marick#82 ] now reports the first failure's normal report
Alex Baranosky
committed
[Issue marick#82 ] formula can now take an optional docstring
Alex Baranosky
committed
[Issue marick#82 ] first version of formula: only reports 0-1 times per formula, regardless of # of generated fact runs
Alex Baranosky
committed
[Issue marick#82 ] removed unused code
Alex Baranosky
committed
[Issue marick#82 ] adding backlog file for the formula epic
Alex Baranosky
committed
[Issue marick#82 ] now reports the first failure's normal report
Alex Baranosky
committed
[Issue marick#82 ] formula can now take an optional docstring
Alex Baranosky
committed
[Issue marick#82 ] updating formulas backlog
Alex Baranosky
committed
Merge branch 'formulas' of github.com:marick/Midje into formulas
Alex Baranosky
committed
[Issue marick#82 ] no longer counts failures as 2 fact runs.
Alex Baranosky
committed

Commits on Feb 22, 2012

[Issue marick#82 ] work in progress
Alex Baranosky
committed
Merge branch 'master' into formulas
Alex Baranosky
committed

Commits on Feb 23, 2012

Merge branch 'master' into formulas
Alex Baranosky
committed

Commits on Feb 25, 2012

Revert "[Issue marick#82 ] work in progress"
Alex Baranosky
committed
workin progress to get formulas to be mergeable into master ready.
Alex Baranosky
committed
[Issue marick#82 ] fixed formulas so that they evaluate their generators, and bodies 100 times each
Alex Baranosky
committed
[Issue marick#82 ] adjustments after checking vs all versions of clojure
Alex Baranosky
committed
[Issue marick#82 ] more ideas added to the formula backlog
Alex Baranosky
committed
[Issue marick#82 ] rollback dynamic bindability of number of fact generations per formula until I can cover it in test properly.
Alex Baranosky
committed
Merge branch 'master' into formulas
Alex Baranosky
committed
added an UNRELEASED section to HISTORY.md - I've been adding to this file after every story. now it is clear which features are released and which are not.
Alex Baranosky
committed
extract util fn to handle pulling optional docstrings from macros' arglists
Alex Baranosky
committed
created midje.util ns for code that helps test private vars
Alex Baranosky
committed
updating HISTORY.md after deploy to clojars
Alex Baranosky
committed

Commits on Feb 26, 2012

[Issue marick#75 ] first step: batch all incorrect call count errors in one aggregated report
Alex Baranosky
committed
[Issue marick#75 ] changing report wording to be match the style shown on the issue.
Alex Baranosky
committed
[Issue marick#75 ] added the expected times called to the report
Alex Baranosky
committed
fixed the doc-string for formula
Alex Baranosky
committed

Commits on Feb 27, 2012

chopped some code off of rotations
Alex Baranosky
committed
removed the colons from FAIL: and WORK TO DO: in the report
Alex Baranosky
committed
now if you expect :times 10 but there are no calls to that fake, then it will say "[expected :times 10, actually never called]" previously said "[expected at least once, actually never called]"
Alex Baranosky
committed

Commits on Feb 29, 2012

[Issue marick#82 ] adding syntax validations for formula macro
Alex Baranosky
committed
[Issue marick#82 ] syntax error if less than 2 elements in the binding
Alex Baranosky
committed
[Issue marick#82 ] generations per formula is now dynamically rebindable
Alex Baranosky
committed
upgrading compatibility test files
Alex Baranosky
committed
[Issue marick#82 ] improving documentation and the backlog
Alex Baranosky
committed

Commits on Mar 1, 2012

[Issue marick#82 ] validates that num generations var is bound to number >= 2
Alex Baranosky
committed
[Issue marick#82 ] removed test.generative -- its not 1.2.1 compatible
Alex Baranosky
committed
[Issue marick#82 ] refactored such that last fact done in formula is just a signal fact to signal the end of the formula, and holds no other purpose
Alex Baranosky
committed

Commits on Mar 2, 2012

[Issue marick#82 ] updated backlog with new insights
Alex Baranosky
committed
[Issue marick#82 ] once a formula detects a failure, won't run any more generated cases
Alex Baranosky
committed
[Issue marick#82 ] updated backlog; organized t-formula
Alex Baranosky
committed
[Issue marick#82 ] formulas can eb set to 1 or more generations now, not 2 or more
Alex Baranosky
committed

Commits on Mar 3, 2012

[Issue marick#82 ] try block is a little more stable
Alex Baranosky
committed
[Issue marick#82 ] updating backlog
Alex Baranosky
committed

Commits on Mar 4, 2012

Commits on Mar 5, 2012

cleanups. Found another use for def-many-methods
Alex Baranosky
committed
cleanups. Found some more uses for def-many-methods
Alex Baranosky
committed

Commits on Mar 6, 2012

assorted cleanups, and added test cases
Alex Baranosky
committed

Commits on Mar 8, 2012

work in progress
Alex Baranosky
committed
Added unfinished message work to UNRELEASED section of HISTORY.md
AlexBaranosky
committed

Commits on Mar 9, 2012

[Issue marick#82 ] updated formula backlog
Alex Baranosky
committed
[Issue marick#82 ] added fact covering simple use of provided in a formulas
Alex Baranosky
committed
[Issue marick#82 ] bunch of partial work
Alex Baranosky
committed

Commits on Mar 10, 2012

my first fnil -- and it feels GOOD.
Alex Baranosky
committed
upgraded kibit, and found a couple little kibbles and bits to tweak.
Alex Baranosky
committed
upgraded 1.4 version to beta4; updated HISTORY to reflect that recent unfinished changes are deployed to clojars in 1.3.2-SNAPSHOT
Alex Baranosky
committed
removed unneeded facts, and cleaned them a bit
Alex Baranosky
committed
work in progress
Alex Baranosky
committed
[Issue marick#82 ] updated formula backlog
Alex Baranosky
committed
[Issue marick#82 ] added fact covering simple use of provided in a formulas
Alex Baranosky
committed
[Issue marick#82 ] bunch of partial work
Alex Baranosky
committed

Commits on Mar 11, 2012

[Issue marick#82 ] dynamic binding of *num-generations-per-formula* works again
Alex Baranosky
committed
[Issue marick#82 ] fixing shrinking to happen at runtime not macroexpand time
Alex Baranosky
committed
[Issue marick#82 ] pulled duplicated code into private fn
Alex Baranosky
committed
[Issue marick#82 ] formulas only allow one check per formula
Alex Baranosky
committed

Commits on Mar 13, 2012

[Issue marick#82 ] formulas can have multiple bindings that each can be shrunken
Alex Baranosky
committed
[Issue marick#82 ] updates to formula backlog
Alex Baranosky
committed
cleaning up tests of syntax validations by making a testing macro that check that each row has the same kind of syntax validation error
Alex Baranosky
committed
[Issue marick#82 ] arrows in against-background in a formula are ignored when looking for check arrows
Alex Baranosky
committed

Commits on Mar 14, 2012

[Issue marick#82 ] a lot more syntax validations for formula
Alex Baranosky
committed

Commits on Mar 15, 2012

[Issue marick#82 ] added notion of a future-formula/pending-formula/incipient-formula/anti...-formula
Alex Baranosky
committed
renames and cleanups etc.
Alex Baranosky
committed
[Issue marick#82 ] added opt-map to formulas with :num-trials key to specify number of trials on a per formula basis
Alex Baranosky
committed
[Issue marick#82 ] renamed *num-generations-per-formula* -> *num-trials*
Alex Baranosky
committed
[Issue marick#82 ] validate that only valid opt-map keys is :num-trials
Alex Baranosky
committed
[Issue marick#82 ] validate that :num-trials can only be set to 1+
Alex Baranosky
committed
[Issue marick#82 ] added one more validation case taking the opts-map into account
Alex Baranosky
committed
[Issue marick#82 ] added opts-map? to :arglists and docstring
Alex Baranosky
committed
[Issue marick#82 ] 'or' makes for cleaner code
Alex Baranosky
committed
[Issue marick#82 ] better test case
Alex Baranosky
committed
Added future-fact with a =streams=> regression.
marick
committed

Commits on Mar 17, 2012

[Issue marick#82 ] use valid-let over when-valid
Alex Baranosky
committed
[Issue marick#82 ] clarifying renames
Alex Baranosky
committed

Commits on Mar 18, 2012

[Issue marick#82 ] clarifying renames - part baznex#2
Alex Baranosky
committed
[Issue marick#82 ] introduce fn pop-opts-map
Alex Baranosky
committed
corrected doc-string
Alex Baranosky
committed
pulled out higher order pop-if fn to use in pop-docstring and pop-opts-map
Alex Baranosky
committed
valid-let over when-valid + let
Alex Baranosky
committed
A more complete lazy streams implementation. Imperfectly tidy.
marick
committed

Commits on Mar 20, 2012

fixed metadata issue with midje.util's functions
Dmitri Naumov
committed
Merge pull request marick#116 from dnaumov/master
AlexBaranosky
committed
thought that format was easier to read than a long str concatted string
Alex Baranosky
committed

Commits on Mar 21, 2012

[Issue marick#82 ] with-num-trials provides a nice syntax for multiple formulas to generate n trials each
Alex Baranosky
committed
cleaning up metaconstant code
Alex Baranosky
committed

Commits on Mar 22, 2012

Update README.md
AlexBaranosky
committed

Commits on Mar 24, 2012

Commits on Mar 26, 2012

[Issue marick#82 ] min length of all shrunken cases seqs determines how many shrunken cases tried
Alex Baranosky
committed
[Issue marick#82 ] missed a comment
Alex Baranosky
committed

Commits on Mar 28, 2012

used set-m monad as essentially a more efficient for expression that stays in sets instead of going back and forth from seq to set. -- I admit, this was fun, and possibly not worth the mental bur…
Alex Baranosky
committed

Commits on Mar 31, 2012

[Issue marick#119 ] moved *expect-checking-fn* to semi-sweet.
Alex Baranosky
committed
assorted validation-error cleanups
Alex Baranosky
committed
removed valid-let in favor of using domonad syntax-validate-m directly. Trying to make eroor-handling clearer by removing indirection
Alex Baranosky
committed
wrote when-valid in terms of (domonad syntax-validate-m ...)
Alex Baranosky
committed
renamed to be the report FORM, since it reutrns a thunked form, it doesn't actually call the report yet
Alex Baranosky
committed
removed tests of private fn that are covered elsewhere
Alex Baranosky
committed
cleanups
Alex Baranosky
committed
renames/cleanups
Alex Baranosky
committed
cleanup of fakes.clj... reducing the private function load, to reduce indirection, to increase my understanding of the code.
Alex Baranosky
committed
namespace qualified a keyword
Alex Baranosky
committed
removed unneeded :testable metadata
Alex Baranosky
committed

Commits on Apr 1, 2012

renamed fake key :lhs to :var and inlined private fn
Alex Baranosky
committed
cleanups
Alex Baranosky
committed
renames, which enable removing a small function
Alex Baranosky
committed
simplified best-call-action finding code
Alex Baranosky
committed
more cleaning renaming and condensing of fakes.clj
Alex Baranosky
committed
missed this rename from last commit
Alex Baranosky
committed
more cleanups
Alex Baranosky
committed
cleanups
Alex Baranosky
committed
added failing future-fact around the thunked =streams=> work
Alex Baranosky
committed

Commits on Apr 2, 2012

upgraded to Kibit 0.0.3
Alex Baranosky
committed

Commits on Apr 3, 2012

reformatted ns macro calls
Alex Baranosky
committed

Commits on Apr 4, 2012

use only the monad... no need for when-valid too
Alex Baranosky
committed
no longer deconstructing tables twice.
Alex Baranosky
committed
removed all the squares
Alex Baranosky
committed
don't need ;; -*- indent-tabs-mode: nil -*- any longer!
Alex Baranosky
committed
dropped the syntax- prefix to validation stuff
Alex Baranosky
committed

Commits on Apr 5, 2012

comments; core.match; cleanups
Alex Baranosky
committed
simpler rotations
Alex Baranosky
committed

Commits on Apr 9, 2012

created reporting package for core reporting code and for various alternate reporters
Alex Baranosky
committed
[Issue marick#121 ] added mostly-empty empty namespace for junit xml report format
Alex Baranosky
committed

Commits on Apr 12, 2012

cleanups for readability
Alex Baranosky
committed
Don't need 'is-' prefix on preds
Alex Baranosky
committed
making against-background wrapping code easier to follow
Alex Baranosky
committed

Commits on Apr 13, 2012

[Issue marick#121 ] created report configuration map to use for reporting fact failures and summaries
Alex Baranosky
committed
Doc-strings are no longer ignored, but are instead reported.
AlexBaranosky
committed

Commits on Apr 14, 2012

[Issue marick#121 ] added a map of formatters
Alex Baranosky
committed

Commits on Apr 15, 2012

upgrade to clojure 1.4.0
Alex Baranosky
committed
added clojure 1.5.0-SNAPSHOT to :multi-deps, to test against
Alex Baranosky
committed

Commits on Apr 20, 2012

added checker combinators: every-checker and some-checker
Alex Baranosky
committed
Document that all regexps must match in throws
marick
committed

Commits on Apr 21, 2012

Commits on Apr 22, 2012

throws checker obeys Midje's extended-false?
marick
committed
data-laden-falsehoods keep the metadata of the value they are created from
Alex Baranosky
committed
added linter, Eastwood, and used it to mark unused fn args with _'s and to fix a erroneous docstring
Alex Baranosky
committed
fixed this non-backwards compatible fn call
Alex Baranosky
committed
marked some unused params with underscores
Alex Baranosky
committed
removed a couple of reflection warnings that snuck back
Alex Baranosky
committed
checker combinators work w/ extended-false? and thus chatty-checkers
Alex Baranosky
committed
lookup fact report fn later, now rebinding formatter works correctly
Alex Baranosky
committed

Commits on Apr 23, 2012

[Issue marick#121 ] removed unnecessary test clutter
Alex Baranosky
committed
moved extended-fn? to util ns
Alex Baranosky
committed
cleanup code to validate seqs of forms
Alex Baranosky
committed
can use lein midje --lazytest from time to time now
Alex Baranosky
committed
Added lazytest repo
marick
committed
Document how to write checkers (including chatty checkers)
marick
committed
and voila! Reporting works once again.
Alex Baranosky
committed
[Issue marick#98 ] stop thunking the rhs of =streams=>
Alex Baranosky
committed

Commits on Apr 25, 2012

Bump version to 1.4.0-SNAPSHOT
marick
committed
Issue: marick#76 fixed by killing core.match in this file.
marick
committed
[Issue marick#76 ] removed use of core.match
Alex Baranosky
committed

Commits on Apr 26, 2012

Commits on Apr 29, 2012

use condp instead of lookup in a map
Alex Baranosky
committed
Revert "use condp instead of lookup in a map" I swear this passed the tests locally :S
Alex Baranosky
committed
kept dispatch fn 'open' by using def-many-methods
Alex Baranosky
committed
use condp instead of map lookup - attempt 2
Alex Baranosky
committed

Commits on May 5, 2012

Commits on May 9, 2012

bump dev version to RC1
marick
committed

Commits on May 10, 2012

starting to think through replacement for background
marick
committed

Commits on May 11, 2012

Commits on Jun 15, 2012

Commits on Jul 16, 2012

Attempt to merge cljs branch w/ current midje
charles-dyfis-net
committed