The files in this .ZIP file are represent two-record types extracted from 
the SEDRA database. For additional information on the structure of the 
Database, see
   G. Kiraz, `Automatic Concordance Generation of Syriac Texts'. 
   In VI Symposium Syriacum 1992, ed. R. Lavenant, Orientalia Christiana
   Analecta 247, Rome, 1994.
******************************************************************

*** NOTE: All references to FEATURRE.TXT files here are to tblWords.txt. (In other words, tblWords.txt is the same as FEATURERE.txt)

The .ZIP file contains two files:
   BFBSREC.TXT  - The text of the Syriac New Testament according to the
                  British and Foreign Bible Society's Edition
   FEATURRE.TXT - A list of all the words in BFBSREC.TXT with morphological
                  information.

BFBSREC.TXT
============
Each line represents a word in the New Testament and contains the following 
fields:
   1. Database address of the entry, e.g. 0:1, 0:2, 0:3. 
      Database address are always represented by 2 numbers in the form:
         FILE_NUMBER:LINE_NUMBER
      Here FILE_NUMBER is always 0.
   2. Word reference, e.g. 520100101
      - The left 2 digits represent the book (52=Matt, 53=Mark, 54=Luke, etc.)
      - Next 2 digits = chapter
      - Next 3 digits = verse
      - Next 2 digits = word
      E.g. 520100101 = Matt, Chapter 1, Verse 1, Word 1
   3. Word address in FEATURRE.TXT, e.g. 33565194d
      First convert the number into hex, e.g. 33565194d=02002A0Ah. 
      The two most singnificant bytes are always 02 which represent the
      FILE_NUMBER. The remaining bytes (i.e. 002A0Ah) give LINE_NUMBER
      in FEATURRE.TXT of the word in question. You have to convert this
      number back into decimal (e.g. 002A0Ah=10762d).
      Now the first word in Matt 1.1 is the one whose address in 
      in FEATURRE.TXT is 02:10762. If you look at line 02:10762 in
      FEATURERE>TXT, you will find the word "CTBA" which is the first word in 
      the Gospel of Matthew!
   4. Attributes. You can ignore this field (or see Kiraz 1994).

FEATURRE.TXT
============
This file is a list of all the words in the Syriac New Testament.
Each line contains the following fields:
   1. Database address of the entry, e.g. 2:1, 2:2, 2:3.
      Database address are always represented by 2 numbers in the form:
         File_Number:Line_Number
      Here FILE_NUMBER is always 2.
   2. Consonantal Word. The word in question (only consonants).
   3. The same word with vowels and diacritic marks.
   4. Morphological Features: A long integer (32 bits). The bits are
      as follows:
      Bit  Meaning
      ---  -------
      0-1  RESEARVED
      2-3  SUFFIX GENDER
           00      SUF_COMMON or <NONE>
           01      SUF_MASCULINE
           10      SUF_FEMININE
           11      <RESERVED>
      4-5  SUFFIX PERSON
           00      <NONE>
           01      THIRD
           10      SECOND
           11      FIRST
      6    SUFFIX NUMBER
           0       SUF_SINGULAR or <NONE>
           1       SUF_PLURAL
      7-8   SUFFIX/CONTRACTION
           00      <NONE>
           01      SUFFIX
           10      CONTRACTION
           11      <RESERVED>
      9-14 PREFIX CODE (0-63)
      15-16 Word GENDER
           00      <NONE>
           01      COMMON
           10      MASCULINE
           11      FEMININE
      17-18 Word PERSON
           00      <NONE>
           01      THIRD
           10      SECOND
           11      FIRST
      19-20 Word NUMBER
           00      <NONE>
           01      SINGULAR
           10      PLURAL
           11      <RESERVED>
      21-22 Nominal STATE
           00      <NONE>
           01      ABSOLUTE
           10      CONSTRUCT
           11      EMPHATIC
      23-25 Verbal TENSE
           000     <NONE>
           001     PERFECT
           010     IMPERFECT
           011     IMPERATIVE
           100     INFINITIVE
           101     ACTIVE_PARTICIPLE
           110     PASSIVE_PARTICIPLE
           111     PARTICIPLES
      26-31 Verbal CONJUGATION
           000000  <NONE>
           000001  PEAL
           000010  ETHPEAL
           000011  PAEL
           000100  ETHPAEL
           000101  APHEL
           000110  ETTAPHAL
           000111  SHAPHEL
           001000  ESHTAPHAL
           001001  SAPHEL
           001010  ESTAPHAL
           001011  PAUEL
           001100  ETHPAUAL
           001101  PAIEL
           001110  ETHPAIAL
           001111  PALPAL
           010000  ETHPALPAL
           010001  PALPEL
           010010  ETHPALPAL
           010011  PAMEL
           010100  ETHPAMAL
           010101  PAREL
           010110  ETHPARAL
           010111  PALI
           011000  ETHPALI
           011001  PAHLI
           011010  ETHPAHLI
           011011  TAPHEL
           011100  ETHAPHAL
           011101-111111 <RESERVED>
      For example, the first line in the file has the field 557056d or
      00088000h or in binary:
         bit 31       Num  Gen                 bit 0
         |            | |  | |                 |
         0000 0000 0000 1000 1000 0000 0000 0000
         bits 15-16 are 01, i.e. Gender = common
         bits 19-20 are 01, i.e. number = singular
5. You can ignore the last field (or see Kiraz 1994).

