comment the phylo.hs file

src/Gargantext/Core/Viz/Phylo.hs

@@ -41,70 +41,114 @@ import Gargantext.Core.Utils.Prefix (unPrefixSwagger)
 import Gargantext.Prelude
 import qualified Data.Text.Lazy as TextLazy
 
+
 ---------------------
 -- | PhyloConfig | --
 ---------------------
 
+
+-- | CorpusParser : control which csv collumns should be taken into account for reconstructing a phylo
 data CorpusParser =
-      Wos  {_wos_limit  :: Int}
-    | Csv  {_csv_limit  :: Int}
-    | Csv' {_csv'_limit :: Int}
+      Wos  
+      -- not used anymore
+      {_wos_limit  :: Int}
+    | Csv  
+      -- consider Publication_Day, Publication_Month, Publication_Year, Authors, Title, Abstract
+      {_csv_limit  :: Int}
+    | Csv' 
+      -- consider Publication_Day, Publication_Month, Publication_Year, Authors, Title, Abstract, Source, Weight
+      {_csv'_limit :: Int}
     deriving (Show,Generic,Eq)
 
 instance ToSchema CorpusParser where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "_")
 
 
-data ListParser = V3 | V4 deriving (Show,Generic,Eq)
+-- | ListParser : is the map list in Csv or in Json?
+data ListParser = 
+      V3 
+      -- the map list is in Csv
+    | V4 
+      -- the map list in in Json
+    deriving (Show,Generic,Eq)
+
 instance ToSchema ListParser
 
 
+-- | SeaElevation : for a given level of observation, 
+--                  define a set of similarity values that will be tested by the sea level rise algorithm
 data SeaElevation =
       Constante
+      -- test a constant set of values (see Gargantext.Core.Viz.Phylo.PhyloMaker.constSeaLadder)      
       { _cons_start :: Double
       , _cons_gap   :: Double }
     | Adaptative
+      -- test a set of values that matches the similarity spectrum of the corpus 
+      -- (see Gargantext.Core.Viz.Phylo.PhyloMaker.constSeaLadder)    
       { _adap_steps :: Double }
     | Evolving
+    -- test a set of values that tries to directly maximize the quality of the phylo
+    -- the similarity spectrum of the corpus (see Gargantext.Core.Viz.Phylo.PhyloMaker.evolvSeaLadder)    
       { _evol_neighborhood :: Bool }      
     deriving (Show,Generic,Eq)
 
 instance ToSchema SeaElevation
 
+
+-- | PhyloSimilarity : define the similarity measure used to for inter temporal matching
 data PhyloSimilarity =
       WeightedLogJaccard
+      -- the default one (see Gargantext.Core.Viz.Phylo.TemporalMatching.weightedLogJaccard')
       { _wlj_sensibility     :: Double
       , _wlj_minSharedNgrams :: Int }
     | WeightedLogSim
+    -- not used
       { _wls_sensibility     :: Double
       , _wls_minSharedNgrams :: Int }
     | Hamming 
+    -- not implemented    
       { _hmg_sensibility     :: Double 
       , _hmg_minSharedNgrams :: Int}
-
     deriving (Show,Generic,Eq)
 
 instance ToSchema PhyloSimilarity where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "")
 
 
-data SynchronyScope = SingleBranch | SiblingBranches | AllBranches
+-- | SynchronyScope : define which groups should be considered by the synchonic clustering
+data SynchronyScope = 
+      SingleBranch
+      -- consider only the groups belonging to the same branch 
+    | SiblingBranches
+      -- consider only the groups belonging to sibling branches 
+      -- (ie. branches that split at the same level of similarity)
+    | AllBranches
+      -- consider every groups of every branches
   deriving (Show,Generic,Eq, ToSchema)
 
-data SynchronyStrategy = MergeRegularGroups | MergeAllGroups
+
+-- | SynchronyStrategy : define which groups should be merged when satisfying the synchonic clustering
+data SynchronyStrategy = 
+      MergeRegularGroups
+      -- only merge groups that don't contain emerging or declining ngrams
+    | MergeAllGroups
+      -- merge every groups
   deriving (Show,Generic,Eq)
 
 instance ToSchema SynchronyStrategy where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "")
 
 
+-- | Synchrony : define the synchronic clustering strategy 
 data Synchrony =
       ByProximityThreshold
+      -- select all groups that satisfy a given similarity threshold
       { _bpt_threshold :: Double
       , _bpt_sensibility :: Double
       , _bpt_scope :: SynchronyScope
       , _bpt_strategy :: SynchronyStrategy }
     | ByProximityDistribution
+    -- select the top x groups sorted by similarity
       { _bpd_sensibility :: Double
       , _bpd_strategy :: SynchronyStrategy }
     deriving (Show,Generic,Eq)
@@ -113,7 +157,10 @@ instance ToSchema Synchrony where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "_")
 
 
-
+-- | TimeUnit : define the temporal granularity
+-- period = size of a PhyloPeriod
+-- step   = step between two subsequent PhyloPeriod
+-- frame  = maximum number of PhyloPeriod considered for intertemporal matching
 data TimeUnit =
       Epoch
       { _epoch_period :: Int
@@ -141,18 +188,28 @@ instance ToSchema TimeUnit where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "")
 
 
-data MaxCliqueFilter = ByThreshold | ByNeighbours deriving (Show,Generic,Eq)
+-- | MaxCliqueFilter : define a strategy for computing MaxClique
+data MaxCliqueFilter = 
+      ByThreshold
+      -- consider ngrams whose confidence probability satisfies a given threshold  
+    | ByNeighbours 
+      -- consider the top x ngrams sorted by confidence probability
+    deriving (Show,Generic,Eq)
 
 instance ToSchema MaxCliqueFilter where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "")
 
 
-
+-- | Cluster : define a ngrams clustering method for computing PhyloGroups
+-- Reference : Uno, Takeaki et al. “LCM ver. 2: Efficient Mining Algorithms for Frequent/Closed/Maximal Itemsets.” 
+--             Workshop on Frequent Itemset Mining Implementations (2004).
 data Cluster =
       Fis
+      -- frequent item set can be filtered by support and size
       { _fis_support :: Int
       , _fis_size    :: Int }
     | MaxClique
+      -- max clique can be filtered by size and threshold
       { _mcl_size      :: Int
       , _mcl_threshold :: Double
       , _mcl_filter    :: MaxCliqueFilter }
@@ -162,15 +219,21 @@ instance ToSchema Cluster where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "")
 
 
+-- | Quality : define a level of observation
 data Quality =
-     Quality { _qua_granularity :: Double
-             , _qua_minBranch   :: Int }
+      Quality 
+      -- _qua_granularity <=> level of observation or λ in (chavalarias, lobbe & delanoe 2021) 
+      -- if λ = 0 then we have one big branch
+      -- if λ = 1 then we have many little branches
+      { _qua_granularity :: Double
+      , _qua_minBranch   :: Int }
       deriving (Show,Generic,Eq)
 
 instance ToSchema Quality where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "_qua_")
 
 
+-- | PhyloConfig : full list of parameters used to reconstruct a Phylomemy from the command line
 data PhyloConfig =
      PhyloConfig { corpusPath     :: FilePath
             , listPath       :: FilePath
@@ -197,6 +260,8 @@ data PhyloConfig =
 -- | SubConfig API & 1Click | --
 --------------------------------
 
+
+-- | PhyloSubConfigAPI : selected list of parameters used to reconstruct a Phylomemy from the API
 data PhyloSubConfigAPI =
   PhyloSubConfigAPI { _sc_phyloProximity :: Double
                     , _sc_phyloSynchrony :: Double
@@ -223,6 +288,8 @@ subConfigAPI2config subConfig = defaultConfig
 -- | SubConfig 1Click | --
 --------------------------  
 
+
+-- | defaultConfig : default configuration used by the 1'Click feature
 defaultConfig :: PhyloConfig
 defaultConfig =
      PhyloConfig { corpusPath = "corpus.csv" -- useful for commandline only
@@ -304,7 +371,7 @@ instance FromJSON Quality
 instance ToJSON Quality
 
 
--- | Software parameters
+-- | Software : software parameters
 data Software =
      Software { _software_name    :: Text
               , _software_version :: Text
@@ -314,14 +381,13 @@ instance ToSchema Software where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "_software_")
 
 
-
 defaultSoftware :: Software
 defaultSoftware =
       Software { _software_name    = pack "GarganText"
                , _software_version = pack "v5" }
 
 
--- | Global parameters of a Phylo
+-- | PhyloParam : global parameters of a Phylo
 data PhyloParam =
      PhyloParam { _phyloParam_version  :: Text
                 , _phyloParam_software :: Software
@@ -332,7 +398,6 @@ instance ToSchema PhyloParam where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "_phyloParam_")
 
 
-
 defaultPhyloParam :: PhyloParam
 defaultPhyloParam =
       PhyloParam { _phyloParam_version  = pack "v3"
@@ -353,14 +418,18 @@ type DateStr = Text
 -- | Ngrams : a contiguous sequence of n terms
 type Ngrams = Text
 
--- Document : a piece of Text linked to a Date
--- date = computational date; date' = original string date yyyy-mm-dd
--- Export Database to Document
+-- | Document : a piece of Text linked to a Date
 data Document = Document
-      { date    :: Date   -- datatype Date {unDate :: Int}
-      , date'   :: DateStr   -- show date
+      { date    :: Date   
+      -- the Int date used to compute the periods, groups, etc. 
+      -- created by toPhyloDate in Gargantext.Core.Viz.Phylo.API.Tools
+      , date'   :: DateStr
+      -- the original String date (yyyy-mm-dd) that will be displayed in the interface
+      -- created by toPhyloDate' in Gargantext.Core.Viz.Phylo.API.Tools
       , text    :: [Ngrams]
       , weight  :: Maybe Double
+      -- a Double attached to each Document that will be used to set up the size of the phylogroup in the interface 
+      -- only taken into account when CorpusParser is CSV'
       , sources :: [Text]
       , docTime :: TimeUnit
       } deriving (Eq,Show,Generic,NFData)
@@ -371,12 +440,14 @@ data Document = Document
 --------------------
 
 
--- | The Foundations of a Phylo created from a given TermList
+-- | PhyloFoundations : store and index all the ngrams (named roots) that will appear in the Phylomemy
 data PhyloFoundations = PhyloFoundations
       { _foundations_roots    :: (Vector Ngrams)
       , _foundations_rootsInGroups :: Map Int [PhyloGroupId] -- map of roots associated to groups
       } deriving (Generic, Show, Eq)
 
+
+-- | PhyloCounts : store various counters related to roots or dates
 data PhyloCounts = PhyloCounts
       { coocByDate    :: !(Map Date Cooc)
       , docsByDate    :: !(Map Date Double)
@@ -386,9 +457,12 @@ data PhyloCounts = PhyloCounts
       , lastRootsFreq :: !(Map Int  Double)
       } deriving (Generic, Show, Eq)
 
+
+-- | PhyloSources : store sources that will be used in the interface to highlight some PhyloGroups
 data PhyloSources = PhyloSources
       { _sources :: !(Vector Text) } deriving (Generic, Show, Eq)
 
+
 instance ToSchema PhyloFoundations where
   declareNamedSchema = genericDeclareNamedSchema (unPrefixSwagger "_foundations_")
 instance ToSchema PhyloCounts where
@@ -521,10 +595,13 @@ type Thr = Double
 
 -- | Pointer : A weighted pointer to a given PhyloGroup
 type Pointer = (PhyloGroupId, Weight)
+
+
 -- | Pointer' : A weighted pointer to a given PhyloGroup with a lower bounded threshold
 type Pointer' = (PhyloGroupId, (Thr,Weight))
 
 data Filiation = ToParents | ToChilds | ToParentsMemory | ToChildsMemory deriving (Generic, Show)
+
 data PointerType = TemporalPointer | ScalePointer deriving (Generic, Show)
 
 
@@ -535,6 +612,8 @@ data PointerType = TemporalPointer | ScalePointer deriving (Generic, Show)
 -- | Support : Number of Documents where a Cluster occurs
 type Support  = Int
 
+
+-- | Clustering : define the structure of a cluster of ngrams
 data Clustering = Clustering
   { _clustering_roots   :: [Int]
   , _clustering_support :: Support