extractor/extract_syntactic.go

   1 /*
   2     extract_syntactic: extract information included in the file
   3     It's based strongly in the syntax of the file.
   4
   5     Copyright (C) 2014 Xavier <quatrilio@gmail.com> and other authors.
   6     See CONTRIBUTORS file.
   7
   8     This program is free software: you can redistribute it and/or modify
   9     it under the terms of the GNU Affero General Public License as
  10     published by the Free Software Foundation, either version 3 of the
  11     License, or (at your option) any later version.
  12
  13     This program is distributed in the hope that it will be useful,
  14     but WITHOUT ANY WARRANTY; without even the implied warranty of
  15     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16     GNU Affero General Public License for more details.
  17
  18     You should have received a copy of the GNU Affero General Public License
  19     along with this program.  If not, see <http://www.gnu.org/licenses/>.
  20
  21     Website: https://github.com/quatrilio/propref
  22
  23 */
  24
  25 /*
  26
  27   extractor is a library for extracting information about mathematical documents.
  28   The information could be: a) syntactic: within the document, like title. b) Meta-informational: typically filesystem information, like size or output from `file` command. c) Contextual: depending of whole documents, as a network [1]. Like the impact index of the document or the popularity of the propositions.
  29
  30   [1] "The information provided by a network itself is different than the sum of the information provided by each individual element in the network" is one of the conclusions from reading the book "Redes Complejas" of Ricard Solé (spanish).
  31
  32 */
  33 package extractor
  34
  35 import (
  36         "regexp"
  37         "log"
  38         "errors"
  39         "github.com/quatrilio/propref/tools"
  40 )
  41
  42 // List of strings we search for in the source documents, by syntax
  43 // BUG(#3): latex commands could have space, return carriages inside .... But only just one return carriage.
  44 // BUG(#5): it does not work with arxiv:1401.5978v1
  45 // BUG(#4): LaTeX: \documentclass{amsart} and others incorporate amsthm (See http://www.ctan.org/pkg/amsart)
  46 var regs map[string]string = map[string]string {
  47         "latex": `(\\usepackage(?:\[\w*(?:,\w*)?\])??{(?:\w*,)?(?P<usepkg_amsmath>amsmath)(?:,\w*)?})`, // for LaTeX
  48 }
  49
  50 // SInfo represents syntactic information of the document. It consists of a common fields of mathematical documents.
  51 type SInfo struct {
  52
  53         Doctype string //Type of the document (latex, context, etc.). Always is lowercase
  54         Subdoctype string //Subtype of the document (for example, amslatex, plain, amsart)
  55         Title string //Title of the document
  56         Authors []Author //list of authors of the document, ordered by their position in the text
  57         Date string //when the document was written. In format "YYYY-MM-DD" (Gregorian calendar)
  58         Resume string //the abstract
  59         License string //the license of the document
  60         MSC []MSC //Mathematical Subject Classification list items
  61         Keywords []string //Most important words of the document
  62         Words []string //Set (not list) of all words
  63         Language string //The language of the document
  64         References []Reference //References appearing in the bibliography
  65         Statements []Statement //List of all statements
  66 }
  67
  68 // Author is related to author information. By now, it's just a string, with his or her complete name (including alias if any).
  69 type Author struct {
  70
  71         Name string //Whole name of the author
  72 }
  73
  74 // MSC represents an Mathematical Subject Classification item.
  75 type MSC struct {
  76
  77         Version uint // the year of revision of the MSC which is used in the document. Eg. 2010
  78         Type string // "primary" or "secondary".
  79         Code string // code. eg "05A10"
  80 }
  81
  82 // Reference implements information of a reference. A reference is basically an information of document which has more information about whatever we said.
  83 type Reference struct {
  84
  85         Author []Author //the authors of the document
  86         Title string //the title of the document
  87         Date string //date which is published or created
  88         URL string //the URL in which we found
  89
  90 }
  91
  92 // Statement implements units of information in mathematical documents: concluding items (like theorems, lemmas, corollaries, etc.) in which we arrive at conclusions about objects, a premise item (like definitions, notation, axioms, hyphotesis, conditions) in which we what we need before reasoning, or a remark item (conjecture, problem, note) in which we just note or remark.
  93 type Statement struct {
  94
  95         Type string //type of the statement
  96         Argument Argument //for propositions only
  97         Assertion string //what is said
  98         Name string //things could have optionally names
  99         References []Reference //if the statements points to a references for better understanding
 100         Documents []Reference //list of documents in which the statement is present
 101
 102 }
 103
 104 // Argument is composed by the hypothesis and the thesis, which are parts of one proposition.
 105 type Argument struct {
 106         Hipothesis string
 107         Thesis string
 108 }
 109
 110 // Extract_Latex extracts SInfo from LaTeX source files.
 111 func Extract_Latex(contents string) (*SInfo, error) {
 112
 113         // Compile and search regular expressions depending on syntax
 114         re, err := regexp.Compile(regs["latex"])
 115
 116         if err != nil {
 117                 log.Fatal(err)
 118                 return nil, errors.New("[PropRef:extract_syntactic:Extract_Latex] " + err.Error())
 119         }
 120
 121         result := SInfo{}
 122         // Extract all the substrings from contents
 123         a := tools.FindStringSubmatchMap(re, contents)
 124         result.Doctype = "latex"
 125         result.Subdoctype = a["usepkg_amsmath"]
 126         return &result, nil
 127
 128
 129 }
 130
 131 // Extract_Syntax extracts information from the string
 132 // according to the specified syntax.
 133 // It retuns a SInfo struct.
 134 func Extract_Syntax(contents string, syntax string) (*SInfo, error) {
 135
 136         switch syntax {
 137                 case "latex": return Extract_Latex(contents)
 138                 default: return nil, errors.New("[PropRef:extract_syntactic] Not suported syntax")
 139         }
 140
 141
 142 }
 143
 144