Merge pull request #28 from poseidon-framework/poseidon_hs_v1403

Updating to Poseidon hs v1403 and new GHC and LTS

.github/workflows/main.yml

@@ -14,7 +14,7 @@ jobs:
     strategy:
       matrix:
         stack: ["latest"]
-        ghc: ["9.2.7"]
+        ghc: ["9.4.7"]
 
     steps:
     # setup and loading cache

CHANGELOG.md

@@ -1,3 +1,4 @@
+- 1.0.0.2: Switched to poseidon-hs v1.4.0.3 and a new compiler (GHC 9.4.7) and stackage snapshot version (LTS-21.17)
 - 1.0.0.1: Switched to poseidon-hs v1.4.0.2
 - 1.0.0.0: Some key updates:
 	- now uses newest poseidon-hs library (v1.4.0.0) with improved and clarified semantics for entity selectors

Dockerfile.linux

@@ -1,12 +1,12 @@
 FROM centos:7
 
 RUN yum -y update
-RUN yum -y install zlib-devel wget ncurses-devel ncurses-compat-libs make gcc
+RUN yum -y install zlib-devel wget ncurses-devel ncurses-compat-libs make gcc gcc-c++
 
 # Install GHC since stack's local install has issues
-RUN wget https://downloads.haskell.org/~ghc/9.2.7/ghc-9.2.7-x86_64-centos7-linux.tar.xz
-RUN tar xvf ghc-9.2.7-x86_64-centos7-linux.tar.xz
-RUN cd ghc-9.2.7; ./configure; make install
+RUN wget https://downloads.haskell.org/~ghc/9.4.7/ghc-9.4.7-x86_64-centos7-linux.tar.xz
+RUN tar xvf ghc-9.4.7-x86_64-centos7-linux.tar.xz
+RUN cd ghc-9.4.7-x86_64-unknown-linux; ./configure; make install
 
 # install stack
 RUN curl -sSL https://get.haskellstack.org/ | sh
@@ -15,5 +15,7 @@ RUN curl -sSL https://get.haskellstack.org/ | sh
 ADD . source
 WORKDIR source
 
-# Install sequenceTools. As described in cryptonite README, cryptonite requires disabling "use_target_attributes"
-RUN stack install --system-ghc --flag cryptonite:-use_target_attributes
+# install
+# - as described in cryptonite README, cryptoniate requires disabling "use_target_attributes"
+# - for bitvec the "simd" flag added in v1.1.5.0 has to be deactivated - the gcc version here does not seem to support it
+RUN stack install --system-ghc --flag cryptonite:-use_target_attributes --flag bitvec:-simd

poseidon-analysis-hs.cabal

@@ -1,5 +1,5 @@
 name:                poseidon-analysis-hs
-version:             1.0.0.1
+version:             1.0.0.2
 synopsis:            A package with analysis-tools to work with Poseidon Genotype Data
 description:         The tools in this package analyse Poseidon-formatted genotype databases, a modular system for storing genotype data from thousands of individuals.
 license:             MIT

src/Poseidon/Analysis/CLI/FStats.hs

@@ -117,13 +117,13 @@ runFstats opts = do
         collectedStats = collectStatSpecGroups statSpecs
         -- new groups can be used on the right hand side of further group definitions, that's why we explicitly exclude them here in the end of the expression
         allEntities = nub (concatMap (map underlyingEntity . snd) groupDefs ++ collectedStats) \\ newGroups
-    checkIfAllEntitiesExist allEntities . getJointIndividualInfo $ allPackages
+    checkIfAllEntitiesExist allEntities =<< getJointIndividualInfo allPackages
 
     -- annotate all individuals in all packages with the new adhoc-group definitions where necessary
     let packagesWithNewGroups = addGroupDefs groupDefs allPackages
 
     -- select only the packages needed for the statistics to be computed
-    relevantPackageNames <- determineRelevantPackages collectedStats . getJointIndividualInfo $ packagesWithNewGroups
+    relevantPackageNames <- determineRelevantPackages collectedStats =<< getJointIndividualInfo packagesWithNewGroups
     let relevantPackages = filter (flip elem relevantPackageNames . posPacNameAndVersion) packagesWithNewGroups
     logInfo $ (show . length $ relevantPackages) ++ " relevant packages for chosen statistics identified:"
     mapM_ (logInfo . show . posPacNameAndVersion) relevantPackages
@@ -230,12 +230,12 @@ type EntityAlleleFreqLookup  = M.Map PoseidonEntity (Maybe Double)
 -- This functioin builds the central Fold that is run over each block of sites of the input data. The return is a tuple of the internal FStats datatypes and the fold.
 buildStatSpecsFold :: (MonadIO m) => [PoseidonPackage] -> [FStatSpec] -> PoseidonIO (FoldM m (EigenstratSnpEntry, GenoLine) BlockData)
 buildStatSpecsFold packages fStatSpecs = do
-    let indInfos = getJointIndividualInfo packages
-        indivNames = map indInfoName indInfos
+    indInfoCollection <- getJointIndividualInfo packages
+    let indivNames = map indInfoName (fst indInfoCollection)
     ploidyVec <- makePloidyVec . getJointJanno $ packages
     entityIndicesLookup <- do
         let collectedSpecs = collectStatSpecGroups fStatSpecs
-        entityIndices <- sequence [resolveUniqueEntityIndices [s] indInfos | s <- collectedSpecs]
+        entityIndices <- sequence [resolveUniqueEntityIndices [s] indInfoCollection | s <- collectedSpecs]
         return . M.fromList . zip collectedSpecs $ entityIndices
     blockAccum <- do
         listOfInnerVectors <- forM fStatSpecs $ \(FStatSpec fType _ _) -> do

src/Poseidon/Analysis/CLI/RAS.hs

@@ -115,13 +115,13 @@ runRAS rasOpts = do
     let newGroups = map (Group . fst) groupDefs
     let allEntities = nub (concatMap (map underlyingEntity . snd) groupDefs ++
             popLefts ++ popRights ++ outgroupSpec) \\ newGroups
-    checkIfAllEntitiesExist allEntities . getJointIndividualInfo $ allPackages
+    checkIfAllEntitiesExist allEntities =<< getJointIndividualInfo allPackages
 
     -- annotate all individuals in all packages with the new adhoc-group definitions where necessary
     let packagesWithNewGroups = addGroupDefs groupDefs allPackages
 
     -- select only the packages needed for the statistics to be computed
-    relevantPackageNames <- determineRelevantPackages (popLefts ++ popRights ++ outgroupSpec) . getJointIndividualInfo $ packagesWithNewGroups
+    relevantPackageNames <- determineRelevantPackages (popLefts ++ popRights ++ outgroupSpec) =<< getJointIndividualInfo packagesWithNewGroups
     let relevantPackages = filter (flip elem relevantPackageNames . posPacNameAndVersion) packagesWithNewGroups
     logInfo $ (show . length $ relevantPackages) ++ " relevant packages for chosen statistics identified:"
     mapM_ (logInfo . show . posPacNameAndVersion) relevantPackages
@@ -246,16 +246,16 @@ readPopConfig fn = do
 
 buildRasFold :: (MonadIO m) => [PoseidonPackage] -> FreqSpec -> FreqSpec -> Double -> Maybe PoseidonEntity -> EntitiesList -> EntitiesList -> PoseidonIO (FoldM m (EigenstratSnpEntry, GenoLine) BlockData)
 buildRasFold packages minFreq maxFreq maxM maybeOutgroup popLefts popRights = do
-    let indInfos = getJointIndividualInfo packages
+    indInfoCollection <- getJointIndividualInfo packages
     ploidyVec <- makePloidyVec . getJointJanno $ packages
     outgroupI <- case maybeOutgroup of
             Nothing -> return []
-            Just o  -> resolveUniqueEntityIndices [o] indInfos
-    leftI <- sequence [resolveUniqueEntityIndices [l] indInfos | l <- popLefts]
-    rightI <- sequence [resolveUniqueEntityIndices [r] indInfos | r <- popRights]
+            Just o  -> resolveUniqueEntityIndices [o] indInfoCollection
+    leftI <- sequence [resolveUniqueEntityIndices [l] indInfoCollection | l <- popLefts]
+    rightI <- sequence [resolveUniqueEntityIndices [r] indInfoCollection | r <- popRights]
     let nL = length popLefts
         nR = length popRights
-    let indivNames = map indInfoName indInfos
+    let indivNames = map indInfoName (fst indInfoCollection)
     return $ FoldM (step ploidyVec indivNames outgroupI leftI rightI) (initialise nL nR) extract
   where
     step :: (MonadIO m) => PloidyVec -> [String] -> [Int] -> [[Int]] -> [[Int]] -> BlockAccumulator ->

stack.yaml

@@ -1,19 +1,18 @@
-resolver: lts-20.24
+resolver: lts-21.17
 
 packages:
 - .
 # - ../poseidon-hs
 
 extra-deps:
   - git: https://github.com/poseidon-framework/poseidon-hs
-    commit: 215e0106dc21b3cbe84287da13e47f19cb7c0e61
+    commit: a760f94e963e171aeb884f855f910d53d92cdef5
   - table-layout-0.9.1.0
   - data-default-instances-base-0.1.0.1
   - sequence-formats-1.7.0
   - pipes-ordered-zip-1.2.1
-  - co-log-0.5.0.0
+  - co-log-0.6.0.2
   - ansi-terminal-0.10.3
   - parsec3-numbers-0.1.0
   - chronos-1.1.5
-  - typerep-map-0.5.0.0
 allow-newer: true

stack.yaml.lock

@@ -5,15 +5,15 @@
 
 packages:
 - completed:
-    commit: 215e0106dc21b3cbe84287da13e47f19cb7c0e61
+    commit: a760f94e963e171aeb884f855f910d53d92cdef5
     git: https://github.com/poseidon-framework/poseidon-hs
     name: poseidon-hs
     pantry-tree:
-      sha256: bf36052db164931ca7e91a158d9c1ec282034a4249160b620ece877d2954b7a7
+      sha256: 0ca7f5e2d425de4f635a32163f1e30bfff23f40218b756f45d26183254607af1
       size: 44449
-    version: 1.4.0.2
+    version: 1.4.0.3
   original:
-    commit: 215e0106dc21b3cbe84287da13e47f19cb7c0e61
+    commit: a760f94e963e171aeb884f855f910d53d92cdef5
     git: https://github.com/poseidon-framework/poseidon-hs
 - completed:
     hackage: table-layout-0.9.1.0@sha256:172b8b3a06c992839a781b77e42bbce689ba87d92c260acb61d187f7e6aad7f2,7154
@@ -44,12 +44,12 @@ packages:
   original:
     hackage: pipes-ordered-zip-1.2.1
 - completed:
-    hackage: co-log-0.5.0.0@sha256:a7e84650eaef7eba2d59ee7664309e79317a7ca67011abedf971f0e6bd6475bb,5448
+    hackage: co-log-0.6.0.2@sha256:91c14447cb1cbdd6d76317e80d2acd75efe4d964975a57f9731d8af56a4fba7a,6395
     pantry-tree:
-      sha256: 33b838c07c8b7e70b2e82bddc889bb1e6386d7e12a9d1593c0b4b263b1fcb925
-      size: 1043
+      sha256: 2ed9bcc839c3681796dd7ef797bcd65cf73e20fc0c542c94cae0a2532db7b3ba
+      size: 1198
   original:
-    hackage: co-log-0.5.0.0
+    hackage: co-log-0.6.0.2
 - completed:
     hackage: ansi-terminal-0.10.3@sha256:e2fbcef5f980dc234c7ad8e2fa433b0e8109132c9e643bc40ea5608cd5697797,3226
     pantry-tree:
@@ -71,16 +71,9 @@ packages:
       size: 581
   original:
     hackage: chronos-1.1.5
-- completed:
-    hackage: typerep-map-0.5.0.0@sha256:34f1ba9b268a6d52e26ae460011a5571e8099b50a3f4a7c8db25dd8efe3be8ee,4667
-    pantry-tree:
-      sha256: ca5565de307d260dc67f6dae0d4d33eee42a3238183461569b5142ceb909c91d
-      size: 1487
-  original:
-    hackage: typerep-map-0.5.0.0
 snapshots:
 - completed:
-    sha256: e019cd29e3f7f9dbad500225829a3f7a50f73c674614f2f452e21bb8bf5d99ea
-    size: 650253
-    url: https://raw.githubusercontent.com/commercialhaskell/stackage-snapshots/master/lts/20/24.yaml
-  original: lts-20.24
+    sha256: 85d2382958c178491d3fe50d770a624621f5ab456beef7d31ac7521f780c9bc7
+    size: 640042
+    url: https://raw.githubusercontent.com/commercialhaskell/stackage-snapshots/master/lts/21/17.yaml
+  original: lts-21.17

test/Poseidon/Analysis/UtilsSpec.hs

@@ -38,7 +38,7 @@ testAddGroupDefs = do
                     ("MyNewGroup2", [Include (Ind "XXX011"), Include (Ind "XXX012")])
                     ]
             let newPacs = addGroupDefs groupDefs pacs
-            let indInfo = getJointIndividualInfo newPacs
+            (indInfo, _) <- getJointIndividualInfo newPacs
             (map indInfoGroups . filter ((=="XXX017") . indInfoName) $ indInfo) `shouldBe` [["POP1", "MyNewGroup1"]]
             (map indInfoGroups . filter ((=="XXX007") . indInfoName) $ indInfo) `shouldBe` [["POP1", "MyNewGroup1"]]
             (map indInfoGroups . filter ((=="XXX011") . indInfoName) $ indInfo) `shouldBe` [["POP1", "MyNewGroup2"]]