joshuaclayton/unused
1
1
Fork 0
mirror of https://github.com/joshuaclayton/unused.git synced 2024-11-04 01:40:25 +03:00
Code Issues Projects Releases Wiki Activity
c5f2a38e80
unused/unused.cabal

96 lines
3.5 KiB
Plaintext
Raw Normal View History

Initial
2016-04-28 12:37:06 +03:00
name: unused
Bump version to 0.1.1.0
2016-05-13 01:11:14 +03:00
version: 0.1.1.0
Update cabal file to reflect proper information
2016-05-11 12:28:01 +03:00
synopsis: A command line tool to identify unused code.
Initial
2016-04-28 12:37:06 +03:00
description: Please see README.md
homepage: https://github.com/joshuaclayton/unused#readme
license: MIT
license-file: LICENSE
author: Josh Clayton
maintainer: sayhi@joshuaclayton.me
copyright: 2016 Josh Clayton
Update cabal file to reflect proper information
2016-05-11 12:28:01 +03:00
category: Development
Initial
2016-04-28 12:37:06 +03:00
build-type: Simple
-- extra-source-files:
cabal-version: >=1.10
Introduce internal yaml configuration of auto low likelihood match handling Why? ==== Handling low likelihood configuration was previously a huge pain, because the syntax in Haskell was fairly terse. This introduces a yaml format internally that ships with the app covering basic cases for Rails, Phoenix, and Haskell. I could imagine getting baselines in here for other languages and frameworks (especially ones I've used and am comfortable with) as a baseline. This also paves the way for searching for user-provided additions and loading those configurations in addition to what we have here.
2016-05-18 18:11:49 +03:00
data-files: data/config.yml
Initial
2016-04-28 12:37:06 +03:00
library
hs-source-dirs: src
Search for terms from a tags file This returns a list of each match, files, and counts based on the current directory.
2016-04-29 00:42:58 +03:00
exposed-modules: Unused.TermSearch
Use regex in ag for simple words Why? ==== ag supports using regular expressions for searches; however, the -Q flag, which was previously always used, resulted in literal search results. By searching literal matches, it would potentially return too many results. For example, with a `me` method in a controller, it'd match words like `awesome` or `method`. This introduces a check where, if the token being searched is only composed of word characters (`[A-Za-z0-9_]`), it'll switch over to use regular expressions with ag and surround the token with non-word matches on either end. The goal here is to reduce false-positives in matches.
2016-05-14 15:04:35 +03:00
, Unused.TermSearch.Internal
Parse results into matches
2016-04-29 05:21:18 +03:00
, Unused.Parser
Extract internal parsing handling to separate module Why? ==== Parsing lines of results was somewhat unreliable, and terms with odd characters were causing problems. This: * extracts parsing into an Unused.Parser.Internal module for ease of testing * fixes cases where certain tokens weren't matching
2016-05-12 20:47:32 +03:00
, Unused.Parser.Internal
Parse results into matches
2016-04-29 05:21:18 +03:00
, Unused.Types
Filter results to matches with one file and one occurrence
2016-04-29 16:58:03 +03:00
, Unused.Util
Initial pass at RemovalLikelihood calculator
2016-05-03 13:41:18 +03:00
, Unused.Regex
Move additional functionality out of Types and into separate modules
2016-05-07 12:56:46 +03:00
, Unused.ResponseFilter
Introduce internal yaml configuration of auto low likelihood match handling Why? ==== Handling low likelihood configuration was previously a huge pain, because the syntax in Haskell was fairly terse. This introduces a yaml format internally that ships with the app covering basic cases for Rails, Phoenix, and Haskell. I could imagine getting baselines in here for other languages and frameworks (especially ones I've used and am comfortable with) as a baseline. This also paves the way for searching for user-provided additions and loading those configurations in addition to what we have here.
2016-05-18 18:11:49 +03:00
, Unused.ResultsClassifier
, Unused.ResultsClassifier.Types
, Unused.ResultsClassifier.Config
Allow search result grouping Why? ==== Grouping results can be helpful to view information differently, e.g. to see highest-offending files or to remove grouping entirely. This introduces a flag to allow overriding the default group (two levels of directory)
2016-05-13 22:57:07 +03:00
, Unused.Grouping
, Unused.Grouping.Internal
, Unused.Grouping.Types
Move additional functionality out of Types and into separate modules
2016-05-07 12:56:46 +03:00
, Unused.LikelihoodCalculator
Read unchanged results from the cache At some point, this also needs to md5 the tags list itself and factor that in (since if the tagging algorithm changes, and new tokens get uncovered, it'd invalidate the cache)
2016-05-15 14:51:54 +03:00
, Unused.Cache
, Unused.Cache.DirectoryFingerprint
Add optparse-applicative for flags This introduces the optparse-applicative library for parsing out any subcommands/flags/args. Currently only supports --no-progress (-P).
2016-05-05 05:55:09 +03:00
, Unused.CLI
, Unused.CLI.Search
, Unused.CLI.SearchError
, Unused.CLI.SearchResult
Better manage column formatting Why? ==== Formatting each column requires context on the column, as well as information on alignment. This extracts the column formatting logic to a specific formatter. ColumnFormatter is coupled to the order of columns/data displayed to the user.
2016-05-09 16:31:06 +03:00
, Unused.CLI.SearchResult.ColumnFormatter
Add optparse-applicative for flags This introduces the optparse-applicative library for parsing out any subcommands/flags/args. Currently only supports --no-progress (-P).
2016-05-05 05:55:09 +03:00
, Unused.CLI.Util
Consolidate progress indicator handling
2016-05-13 15:03:26 +03:00
, Unused.CLI.ProgressIndicator
, Unused.CLI.ProgressIndicator.Internal
, Unused.CLI.ProgressIndicator.Types
Introduce internal yaml configuration of auto low likelihood match handling Why? ==== Handling low likelihood configuration was previously a huge pain, because the syntax in Haskell was fairly terse. This introduces a yaml format internally that ships with the app covering basic cases for Rails, Phoenix, and Haskell. I could imagine getting baselines in here for other languages and frameworks (especially ones I've used and am comfortable with) as a baseline. This also paves the way for searching for user-provided additions and loading those configurations in addition to what we have here.
2016-05-18 18:11:49 +03:00
other-modules: Paths_unused
Initial
2016-04-28 12:37:06 +03:00
build-depends: base >= 4.7 && < 5
Search for terms from a tags file This returns a list of each match, files, and counts based on the current directory.
2016-04-29 00:42:58 +03:00
, process
Parse results into matches
2016-04-29 05:21:18 +03:00
, parsec
Reorganize data structure to handle parsing errors This extracts the previous data structure from groupBy into an actual Data.Map.Strict String [TermMatch], as well as another type (ParseResponse) capturing invalid and valid responses.
2016-04-29 12:28:03 +03:00
, containers
Improve output by grouping by directory structure
2016-04-30 23:05:12 +03:00
, filepath
Read unchanged results from the cache At some point, this also needs to md5 the tags list itself and factor that in (since if the tagging algorithm changes, and new tokens get uncovered, it'd invalidate the cache)
2016-05-15 14:51:54 +03:00
, directory
Initial pass at RemovalLikelihood calculator
2016-05-03 13:41:18 +03:00
, regex-tdfa
Add optparse-applicative for flags This introduces the optparse-applicative library for parsing out any subcommands/flags/args. Currently only supports --no-progress (-P).
2016-05-05 05:55:09 +03:00
, terminal-progress-bar
, ansi-terminal
Hook into interrupt to trigger other behavior Why? ==== Unused hides the cursor and potentially does other things to the window that may leave it in an odd state. This introduces a hook to run any state cleanup, including re-enabling the cursor, when a user sends a SIGINT to the program.
2016-05-10 00:45:42 +03:00
, unix
Parallelize search Why? ==== Searching hundreds or thousands of tokens with ag can be slow; this introduces parallel processing of search so results are returned more quickly.
2016-05-13 21:46:23 +03:00
, parallel-io
Introduce internal yaml configuration of auto low likelihood match handling Why? ==== Handling low likelihood configuration was previously a huge pain, because the syntax in Haskell was fairly terse. This introduces a yaml format internally that ships with the app covering basic cases for Rails, Phoenix, and Haskell. I could imagine getting baselines in here for other languages and frameworks (especially ones I've used and am comfortable with) as a baseline. This also paves the way for searching for user-provided additions and loading those configurations in addition to what we have here.
2016-05-18 18:11:49 +03:00
, yaml
, bytestring
, text
, unordered-containers
Warn on everything, error on anything
2016-05-11 12:32:42 +03:00
ghc-options: -Wall -Werror -O2
Initial
2016-04-28 12:37:06 +03:00
default-language: Haskell2010
executable unused
hs-source-dirs: app
main-is: Main.hs
Run optimizations
2016-05-09 15:29:09 +03:00
ghc-options: -threaded -rtsopts -with-rtsopts=-N -Wall -Werror -O2
Initial
2016-04-28 12:37:06 +03:00
build-depends: base
, unused
Add optparse-applicative for flags This introduces the optparse-applicative library for parsing out any subcommands/flags/args. Currently only supports --no-progress (-P).
2016-05-05 05:55:09 +03:00
, optparse-applicative
Initial
2016-04-28 12:37:06 +03:00
default-language: Haskell2010
test-suite unused-test
type: exitcode-stdio-1.0
hs-source-dirs: test
main-is: Spec.hs
build-depends: base
, unused
Add Hspec and tests around parsing
2016-05-08 13:29:55 +03:00
, hspec
, containers
other-modules: Unused.ParserSpec
Extract internal parsing handling to separate module Why? ==== Parsing lines of results was somewhat unreliable, and terms with odd characters were causing problems. This: * extracts parsing into an Unused.Parser.Internal module for ease of testing * fixes cases where certain tokens weren't matching
2016-05-12 20:47:32 +03:00
, Unused.Parser.InternalSpec
Add Hspec and tests around parsing
2016-05-08 13:29:55 +03:00
, Unused.ResponseFilterSpec
, Unused.TypesSpec
Improve likelihood calculation and include reasons for evaluation Why? ==== A simple calculation ("yes, this should be removed" or "no, this is probably fine") is frankly not enough information for someone evaluating their codebase to understand why we made the decision. This introduces a removal reason, so a user understands why we ranked it the way we did, and adds additional logic around a method and its tests to determine if a method exists and is only being used in the tests (if so, it should probably be deleted). This is done with an Occurrances record, which is created for total files, test code, and non-test code. The test code logic is somewhat naive but works in most cases. It doesn't ensure a particular directory, in the case that tests live alongside source code (e.g. Go), and captures RSpec cases as well.
2016-05-08 15:16:32 +03:00
, Unused.LikelihoodCalculatorSpec
Allow search result grouping Why? ==== Grouping results can be helpful to view information differently, e.g. to see highest-offending files or to remove grouping entirely. This introduces a flag to allow overriding the default group (two levels of directory)
2016-05-13 22:57:07 +03:00
, Unused.Grouping.InternalSpec
Use regex in ag for simple words Why? ==== ag supports using regular expressions for searches; however, the -Q flag, which was previously always used, resulted in literal search results. By searching literal matches, it would potentially return too many results. For example, with a `me` method in a controller, it'd match words like `awesome` or `method`. This introduces a check where, if the token being searched is only composed of word characters (`[A-Za-z0-9_]`), it'll switch over to use regular expressions with ag and surround the token with non-word matches on either end. The goal here is to reduce false-positives in matches.
2016-05-14 15:04:35 +03:00
, Unused.TermSearch.InternalSpec
Test Util.groupBy
2016-05-15 12:53:29 +03:00
, Unused.UtilSpec
Introduce internal yaml configuration of auto low likelihood match handling Why? ==== Handling low likelihood configuration was previously a huge pain, because the syntax in Haskell was fairly terse. This introduces a yaml format internally that ships with the app covering basic cases for Rails, Phoenix, and Haskell. I could imagine getting baselines in here for other languages and frameworks (especially ones I've used and am comfortable with) as a baseline. This also paves the way for searching for user-provided additions and loading those configurations in addition to what we have here.
2016-05-18 18:11:49 +03:00
, Paths_unused
Error on warnings
2016-05-07 13:23:38 +03:00
ghc-options: -threaded -rtsopts -with-rtsopts=-N -Wall -Werror
Initial
2016-04-28 12:37:06 +03:00
default-language: Haskell2010
source-repository head
type: git
location: https://github.com/joshuaclayton/unused
Reference in New Issue Copy Permalink