Class: ValuesReader
- Inherits:
-
Object
- Object
- ValuesReader
- Includes:
- LogUtils::Logging, TextUtils::ValueHelper
- Defined in:
- lib/textutils/reader/values_reader.rb
Instance Method Summary collapse
-
#each_line ⇒ Object
support multi line records.
-
#find_values(line) ⇒ Object
todo: move to helper for reuse a la find_key_n_title ??? use different/better name ?? e.g.
-
#initialize(path, more_attribs = {}) ⇒ ValuesReader
constructor
A new instance of ValuesReader.
Methods included from TextUtils::ValueHelper
#find_grade, #find_key_n_title, #is_address?, #is_region?, #is_taglist?, #is_website?, #is_year?, #match_abv, #match_brewery, #match_city, #match_country, #match_hl, #match_kcal, #match_km_squared, #match_metro, #match_metro_flag, #match_metro_pop, #match_number, #match_og, #match_region_for_country, #match_supra, #match_supra_flag, #match_website, #match_year
Constructor Details
#initialize(path, more_attribs = {}) ⇒ ValuesReader
Returns a new instance of ValuesReader.
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 |
# File 'lib/textutils/reader/values_reader.rb', line 13 def initialize( path, more_attribs={} ) @more_attribs = more_attribs ### workaround/hack # if path includes newline assume it's a string buffer not a file name # fix: use from_file an from_string etc. for ctor # check what is the best convention (follow ???) if path =~ /\n/m @path = 'stringio' # what name to use ??? @data = path.dup # make a duplicate ?? why? why not? else @path = path @data = File.read_utf8( @path ) end end |
Instance Method Details
#each_line ⇒ Object
support multi line records
39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 |
# File 'lib/textutils/reader/values_reader.rb', line 39 def each_line # support multi line records inside_record = false blank_counter = 0 # count of number of blank lines (note: 1+ blank lines clear multi-line record) values = [] # keep track of last header # e.g. lines like # ___________________________________ # - Brauerei Schwechat (Brau Union) # # laster_header will be 'Brauerei Schwechat (Brau Union)' # gets passed along as an attribue e.g. more_attribs[:header]='Brauerei Schwechat (Brau Union)' last_header = nil @data.each_line do |line| ## allow alternative comment lines ## e.g. -- comment or ## % comment ## why? # might get used by markdown for marking headers, for example ## NB: for now alternative comment lines not allowed as end of line style e.g ## some data, more data -- comment here if line =~ /^\s*#/ || line =~ /^\s*--/ || line =~ /^\s*%/ || line =~ /^\s*__/ # skip komments and do NOT copy to result (keep comments secret!) logger.debug 'skipping comment line' next end if line =~ /^\s*$/ # kommentar oder leerzeile überspringen blank_counter += 1 logger.debug "skipping blank line (#{blank_counter})" next end # pass 1) remove possible trailing eol comment ## e.g -> nyc, New York # Sample EOL Comment Here (with or without commas,,,,) ## becomes -> nyc, New York line = line.sub( /\s+#.+$/, '' ) # pass 2) remove leading and trailing whitespace line = line.strip if line =~ /^-\s+/ # check for group headers e.g. - St. James Brewery if values.length > 0 # check if we already processed a record? if yes; yield last record (before reset) attribs, more_values = find_key_n_title( values ) attribs = attribs.merge( @more_attribs ) # e.g. merge country_id and other defaults if present attribs[:header] = last_header unless last_header.nil? # add optional header attrib yield( attribs, more_values ) values = [] end inside_record = false blank_counter = 0 # update last_header last_header = line.gsub( /^-\s/, '' ) # cut-off leading marker and space logger.info " update group header >#{last_header}<" next elsif line =~ /^\[([a-z][a-z]+)\]/ ### check for multiline record ## must start with key e.g. [guiness] ## for now only supports key with letter a-z (no digits/numbers or underscore or dots) if values.length > 0 # check if we already processed a record? if yes; yield last record (before reset) attribs, more_values = find_key_n_title( values ) attribs = attribs.merge( @more_attribs ) # e.g. merge country_id and other defaults if present attribs[:header] = last_header unless last_header.nil? # add optional header attrib yield( attribs, more_values ) values = [] end inside_record = true blank_counter = 0 # NB: every additional line is one value e.g. city:wien, etc. # allows you to use any chars logger.debug " multi-line record w/ key »#{$1}«" values = [$1.dup] # add key as first value in ary elsif inside_record && blank_counter == 0 && line =~ /\/{2}/ # check address line (must contain //) values += [line.dup] # assume single value column (no need to escape commas) elsif inside_record && blank_counter == 0 && line =~ /^[a-z][a-z0-9.]*[a-z0-9]:/ # check key: value pair values += [line.dup] # assume single value column (no need to escape commas) else if inside_record && blank_counter == 0 # continue adding more values values += find_values( line ) else # assume single-line (stand-alone / classic csv) record if values.length > 0 attribs, more_values = find_key_n_title( values ) attribs = attribs.merge( @more_attribs ) # e.g. merge country_id and other defaults if present attribs[:header] = last_header unless last_header.nil? # add optional header attrib yield( attribs, more_values ) values = [] end inside_record = false blank_counter = 0 values = find_values( line ) end end end # each lines # do NOT forget to yield last line (if present/processed) if values.length > 0 attribs, more_values = find_key_n_title( values ) attribs = attribs.merge( @more_attribs ) # e.g. merge country_id and other defaults if present attribs[:header] = last_header unless last_header.nil? # add optional header attrib yield( attribs, more_values ) end end |
#find_values(line) ⇒ Object
todo:
move to helper for reuse a la find_key_n_title ???
use different/better name ?? e.g. find_values_in_line or split_line_into_values ??
164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 |
# File 'lib/textutils/reader/values_reader.rb', line 164 def find_values( line ) ## note returns an array of values (strings) = '«KOMMA»' = '« »' # guard escaped commas # e.g. convert \, to «KOMMA» line = line.gsub( '\,', ) # note: use generic separator (allow us to configure separator) # e.g « » line = line.gsub( ',', ) # restore escaped commas (before split) line = line.gsub( , ',' ) logger.debug "line: |»#{line}«|" values = line.split( ) # pass 1) remove leading and trailing whitespace for values values = values.map { |value| value.strip } ##### todo/fix: # !!!REMOVE!!! # remove support of comment column? (NB: must NOT include commas) # pass 2) remove comment columns # # todo/fix: check if still possible ?? - add an example here how it looks like/works values = values.select do |value| if value =~ /^#/ ## start with # treat it as a comment column; e.g. remove it logger.info " removing column with value »#{value}«" false else true end end logger.debug " values: |»#{values.join('« »')}«|" values end |