csvreader 0.2.0 → 0.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: af0fcea1b598e6123786a05532a6f5b2e10a4095
-  data.tar.gz: ba2dc18a6076e425847b440c05819e898f0a66b2
+  metadata.gz: a9bc6971bd638abc67e8e82e241dbb370602b0d5
+  data.tar.gz: 062f2727188a6f3705c21a5cc825194f84bea41c
 SHA512:
-  metadata.gz: 28f60b98574e5331b53280f27017fae776c787ee1b7a56815c8a8f9c21a0926e6f561ca8a75f1464f1743849f989a52f122fbe4f20086de8159cf2df53b71bbe
-  data.tar.gz: 6d5b80b11e4774bc227bffe62bc829ab19b70f22fd69ca35c54526b85f261fa5c4bf0a7d87c9ba715738f50a6710bdd843f3b6cc1581f0d88744332fdf062796
+  metadata.gz: 595f1c779e0457377fe5c09602cba1ce7754803b35b9280e06dfc752759da441c708db830cb159076ce3f445b3b6aaf1fef459ff9eaace4ae6a436988e52455f
+  data.tar.gz: f4dc02242912ba15bef498838093f85de22c3670b5bcb2b92139adbd9343cbddec753f755729f85b2962df26d724ff069cc73b8f5cccd3edb896dc0d3ac26969

data/README.md

@@ -164,7 +164,7 @@ see [`TabReader` »](https://github.com/datatxt/tabreader).
 
 Two major design bugs and many many minor.
 
-1) The CSV class uses `line.split(`,`)` with some kludges (†) with the claim its faster.
+(1) The CSV class uses `line.split(',')` with some kludges (†) with the claim its faster.
 What?! The right way: CSV needs its own purpose-built parser. There's no other
 way you can handle all the (edge) cases with double quotes and escaped doubled up
 double quotes. Period.
@@ -175,7 +175,7 @@ Or handling double quotes inside values and so on and on.
 
 (†): kludge - a workaround or quick-and-dirty solution that is clumsy, inelegant, inefficient, difficult to extend and hard to maintain
 
-2) The CSV class returns `nil` for `,,` but an empty string (`""`)
+(2) The CSV class returns `nil` for `,,` but an empty string (`""`)
 for `"","",""`. The right way: All values are always strings. Period.
 
 If you want to use `nil` you MUST configure a string (or strings)

data/lib/csvreader/reader.rb

@@ -6,6 +6,34 @@ module Csv    ## check: rename to CsvSettings / CsvPref / CsvGlobals or similar
   ## STD_CSV_ENGINE = CSV   ## to avoid name confusion use longer name - why? why not? find a better name?
   ## use __CSV__ or similar? or just ::CSV ??
 
+
+class Dialect   ## todo: use a module - it's just a namespace/module now - why? why not?
+  ###
+  # (auto-)add these flavors/dialects:
+  #     :tab                   -> uses TabReader(!)
+  #     :strict|:rfc4180
+  #     :unix                   -> uses unix-style escapes e.g. \n \" etc.
+  #     :windows|:excel
+  #     :guess|:auto     -> guess (auto-detect) separator - why? why not?
+
+  ##  e.g. use Dialect.registry[:unix] = { ... } etc.
+  ##   note use @@ - there is only one registry
+  def self.registry() @@registry ||={} end
+
+  ## add built-in dialects:
+  ##    trim - use strip? why? why not? use alias?
+  registry[:tab]     = {}   ##{ class: TabReader }
+  registry[:strict]  = { strict: true, trim: false }   ## add no comments, blank lines, etc. ???
+  registry[:rfc4180] = :strict    ## alternative name
+  registry[:windows] = {}
+  registry[:excel]   = :windows
+  registry[:unix]    = {}
+
+  ## todo: add some more
+end  # class Dialect
+
+
+
   class Configuration
 
     puts "CSV::VERSION:"
@@ -23,6 +51,9 @@ module Csv    ## check: rename to CsvSettings / CsvPref / CsvGlobals or similar
 
 
     attr_accessor :sep   ## col_sep (column separator)
+    attr_accessor :na    ## not available (string or array of strings or nil) - rename to nas/nils/nulls - why? why not?
+    attr_accessor :trim        ### allow ltrim/rtrim/trim - why? why not?
+    attr_accessor :dialect
 
     def initialize
       @sep = ','
@@ -32,6 +63,8 @@ module Csv    ## check: rename to CsvSettings / CsvPref / CsvGlobals or similar
       self  ## return self for chaining
     end
 
+    def trim?() @trim; end   ## strip leading and trailing spaces
+
     def blank?( line )
       ## note:  blank line does NOT include "blank" with spaces only!!
       ##          use BLANK_REGEX in skip_lines to clean-up/skip/remove/ignore
@@ -96,46 +129,53 @@ end   # module Csvv
 
 class CsvReader
 
-  def self.foreach( path, sep: Csv.config.sep, headers: false )
+  def self.parse_line( txt, sep:        Csv.config.sep,
+                            trim:       Csv.config.trim?,
+                            na:         Csv.config.na,
+                            dialect:    Csv.config.dialect,
+                            converters: nil)
+    ## note: do NOT include headers option (otherwise single row gets skipped as first header row :-)
     csv_options = Csv.config.default_options.merge(
-                     headers: headers,
-                     col_sep: sep,
-                     external_encoding: 'utf-8'  ## note:  always (auto-)add utf-8 external encoding for now!!!
+                    headers: false,  ## note: always turn off headers!!!!!!
+                    col_sep: sep
     )
+    ## pp csv_options
+    CSV.parse_line( txt, csv_options )
+  end
 
-    CSV.foreach( path, csv_options ) do |row|
-      yield( row )    ## check/todo: use block.call( row ) ## why? why not?
-    end
+  def self.parse( txt, sep: Csv.config.sep, headers: false )
+    csv_options = Csv.config.default_options.merge(
+                     headers: headers,
+                     col_sep: sep
+    )
+    ## pp csv_options
+    CSV.parse( txt, csv_options )
   end
 
   def self.read( path, sep: Csv.config.sep, headers: false )
     ## note: use our own file.open
     ##   always use utf-8 for now
     ##    check/todo: add skip option bom too - why? why not?
-    txt = File.open( path, 'r:utf-8' )
+    txt = File.open( path, 'r:bom|utf-8' )
     parse( txt, sep: sep, headers: headers )
   end
 
-  def self.parse( txt, sep: Csv.config.sep, headers: false )
+  def self.foreach( path, sep: Csv.config.sep, headers: false )
     csv_options = Csv.config.default_options.merge(
                      headers: headers,
-                     col_sep: sep
+                     col_sep: sep,
+                     external_encoding: 'utf-8'  ## note:  always (auto-)add utf-8 external encoding for now!!!
     )
-    ## pp csv_options
-    CSV.parse( txt, csv_options )
-  end
 
+    ##  todo/check/fix:
+    ##  can use bom e.g. 'bom|utf-8' - how?
+    ##   raises ArgumentError: unknown encoding name - bom|utf-8
 
-  def self.parse_line( txt, sep: Csv.config.sep )
-    ## note: do NOT include headers option (otherwise single row gets skipped as first header row :-)
-    csv_options = Csv.config.default_options.merge(
-                    headers: false,  ## note: always turn off headers!!!!!!
-                    col_sep: sep
-    )
-    ## pp csv_options
-    CSV.parse_line( txt, csv_options )
-  end
 
+    CSV.foreach( path, csv_options ) do |row|
+      yield( row )    ## check/todo: use block.call( row ) ## why? why not?
+    end
+  end
 
   def self.header( path, sep: Csv.config.sep )   ## use header or headers - or use both (with alias)?
       # read first lines (only)
@@ -148,7 +188,7 @@ class CsvReader
       ##  - NOT a blank line
 
       lines = ''
-      File.open( path, 'r:utf-8' ) do |f|
+      File.open( path, 'r:bom|utf-8' ) do |f|
 
         ## todo/fix: how to handle empty files or files without headers?!
 
@@ -171,31 +211,20 @@ class CsvReader
       parse_line( lines, sep: sep )
     end  # method self.header
 
-    ####################
-    # helper methods
-    def self.unwrap( row_or_array )   ## unwrap row - find a better name? why? why not?
-      ## return row values as array of strings
-      if row_or_array.is_a?( CSV::Row )
-        row = row_or_array
-        row.fields   ## gets array of string of field values
-      else  ## assume "classic" array of strings
-        array = row_or_array
-      end
-    end
 end # class CsvReader
 
 
 
 class CsvHashReader
 
-def self.read( path, sep: Csv.config.sep, headers: true )
-  CsvReader.read( path, sep: sep, headers: headers )
-end
-
 def self.parse( txt, sep: Csv.config.sep, headers: true )
   CsvReader.parse( txt, sep: sep, headers: headers )
 end
 
+def self.read( path, sep: Csv.config.sep, headers: true )
+  CsvReader.read( path, sep: sep, headers: headers )
+end
+
 def self.foreach( path, sep: Csv.config.sep, headers: true, &block )
   CsvReader.foreach( path, sep: sep, headers: headers, &block )
 end

data/lib/csvreader/version.rb

@@ -4,7 +4,7 @@
 class CsvReader   ## note: uses a class for now - change to module - why? why not?
 
   MAJOR = 0    ## todo: namespace inside version or something - why? why not??
-  MINOR = 2
+  MINOR = 3
   PATCH = 0
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: csvreader
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.0
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-08-19 00:00:00.000000000 Z
+date: 2018-08-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rdoc