Module Pdftext

module Pdftext: sig .. end

Parsing fonts and extracting text from content streams and PDF strings

Data Types

type type3_glpyhs = {

	`fontbbox : float * float * float * float;`
	`fontmatrix : Pdftransform.transform_matrix;`
	`charprocs : (string * Pdf.pdfobject) list;`
	`type3_resources : Pdf.pdfobject;`

}

type simple_fonttype =

`\|`	`Type1`
`\|`	`MMType1`
`\|`	`Type3 of type3_glpyhs`
`\|`	`Truetype`

type fontfile =

`\|`	`FontFile of int`
`\|`	`FontFile2 of int`
`\|`	`FontFile3 of int`

type fontdescriptor = {

	`ascent : float;`
	`descent : float;`
	`avgwidth : float;`
	`maxwidth : float;`
	`flags : int;`
	`fontbbox : float * float * float * float;`
	`italicangle : float;`
	`capheight : float;`
	`xheight : float;`
	`stemv : float;`
	`fontfile : fontfile option;`
	`charset : string list option;`
	`tounicode : (int, string) Stdlib.Hashtbl.t option;`

}

type differences = (string * int) list

type encoding =

`\|`	`ImplicitInFontFile`
`\|`	`StandardEncoding`
`\|`	`MacRomanEncoding`
`\|`	`WinAnsiEncoding`
`\|`	`MacExpertEncoding`
`\|`	`CustomEncoding of encoding * differences`
`\|`	`FillUndefinedWithStandard of encoding`

type fontmetrics = float array

type simple_font = {

	`fonttype : simple_fonttype;`
	`basefont : string;`
	`firstchar : int;`
	`lastchar : int;`
	`widths : int array;`
	`fontdescriptor : fontdescriptor option;`
	`fontmetrics : fontmetrics option;`
	`encoding : encoding;`

}

type standard_font =

`\|`	`TimesRoman`
`\|`	`TimesBold`
`\|`	`TimesItalic`
`\|`	`TimesBoldItalic`
`\|`	`Helvetica`
`\|`	`HelveticaBold`
`\|`	`HelveticaOblique`
`\|`	`HelveticaBoldOblique`
`\|`	`Courier`
`\|`	`CourierBold`
`\|`	`CourierOblique`
`\|`	`CourierBoldOblique`
`\|`	`Symbol`
`\|`	`ZapfDingbats`

type cid_system_info = {

	`registry : string;`
	`ordering : string;`
	`supplement : int;`

}

type composite_CIDfont = {

	`cid_system_info : cid_system_info;`
	`cid_basefont : string;`
	`cid_fontdescriptor : fontdescriptor;`
	`cid_widths : (int * float) list;`
	`cid_default_width : int;`

}

type cmap_encoding =

`\|`	`Predefined of string`
`\|`	`CMap of int`

type font =

`\|`	`StandardFont of standard_font * encoding`
`\|`	`SimpleFont of simple_font`
`\|`	`CIDKeyedFont of string * composite_CIDfont * cmap_encoding`

String representations of fonts

val string_of_standard_font : standard_font -> string

Returns a string such as "Times-Bold" for Pdftext.TimesBold etc.

val standard_font_of_name : string -> standard_font option

Parses a string such as "/Times-Bold" or "/TimesNewRoman,Bold" to Pdftext.TimesRomanBold etc.

val string_of_font : font -> string

A debug string for the whole font datatype.

Reading a Font

val read_font : Pdf.t -> Pdf.pdfobject -> font

Read a font from a given document and object

Writing a Font

val write_font : ?objnum:int -> Pdf.t -> font -> int

Write a font to a given document, returning the object number for the main font dictionary

Utility functions

val is_unicode : string -> bool

Is a PDF string UTF16be (i.e does it have a byte order marker at the beginning)?

val is_identity_h : font -> bool

Is a font Identity H?

val codepoints_of_utf8 : string -> int list

A list of unicode codepoints for a UTF8 string

val utf8_of_codepoints : int list -> string

A UTF8 string for a list of unicode codepoints

val codepoints_of_utf16be : string -> int list

A list of unicode codepoints for a UTF16BE string

val utf16be_of_codepoints : int list -> string

A UTF16BE string for a list of unicode codepoints (with BOM)

Text from strings outside page content

val utf8_of_pdfdocstring : string -> string

Take a pdf string (which will be either pdfdocencoding or UTF16BE) and return a string representing the same unicode codepoints in UTF8

val pdfdocstring_of_utf8 : string -> string

Take a UTF8 string and convert to pdfdocencoding (if no unicode-only characters are used) or UTF16BE (if they are))

val pdfdocstring_of_codepoints : int list -> string

Build a pdf string in pdfdocencoding (if no unicode-only characters are used) or UTF16BE (if they are)

val codepoints_of_pdfdocstring : string -> int list

Produce a list of unicode codepoints from a pdfdocencoding or UTF16BE pdf document string

val simplify_utf16be : string -> string

Remake a UTF16BE string into a PDFDocEncoding string if all characters are in PDFDocEncoding

Text from strings inside page content

type text_extractor

The type of text extractors.

val text_extractor_of_font : Pdf.t -> Pdf.pdfobject -> text_extractor

Build a text extractor from a document and font object

val text_extractor_of_font_real : font -> text_extractor

Build a text extractor from a document and a font

val codepoints_of_text : text_extractor -> string -> int list

Return a list of unicode points from a given extractor and string (for example from a Pdfpages.Op_Tj or Op_TJ operator).

val glyphnames_of_text : text_extractor -> string -> string list

Return a list of glyph names from a given extractor and string

Building text for strings inside page content

val charcode_extractor_of_font : ?debug:bool -> Pdf.t -> Pdf.pdfobject -> int -> int option

Return the character code for a given unicode codepoint, if it exists in the encoding and font object. If debug is set (default false) missing characters are reported to stderr.

val charcode_extractor_of_font_real : ?debug:bool -> font -> int -> int option

Return the character code for a given unicode codepoint, if it exists in the encoding and font. If debug is set (default false) missing characters are reported to stderr.

val table_of_encoding : encoding -> (int, string) Stdlib.Hashtbl.t

Table of all the entries in an encoding.

val reverse_table_of_encoding : encoding -> (string, int) Stdlib.Hashtbl.t

Reverse table of all the entries in an encoding.

val parse_tounicode : Pdf.t -> Pdf.pdfobject -> (int * string) list

Parse a /ToUnicode entry.