Structured LISTs
Definition of a list given by a .LST text file can be extended by
additional specifications telling about the structure of a list,
i.e. how the text in the list can be interpreted as a sequence
of cases consisting of one or more lines and how various fields
within the text are identified.
In this form a list corresponds to a data file with string and
numeric variables. However, there is neither a fixed length nor
a fixed position of a field in one record or case.
Thus a structured LIST in Survo consists of a sequence of text chapters
in edit fields with a division into cases (records) of one or more
edit lines and within each case of fields (variables) in free order.
Structured LISTs (continued)
The cases are limited either by empty lines or by control characters
in the control column (one character shows where the case begins and
another where it ends).
The fields are indicated in different ways:
1) by characters in the control column,
2) by shadow characters (colours),
3) by key texts within the text itself.
The two first alternatives are most suitable since they admit coding
of textual data without inserting technical notation in the text itself.
It is also possible to define subcases (by the above-mentioned means)
to indicate their range within a case and define fields within
a particular subcase. The subcases can also be nested thus enabling
hierarchical structures in lists.
Also several structured definitions can be given to the same sequence
of edit file chapters by different .LST files.
For the time being strucured lists can be processed by LIST SORT.
Structured LISTs (continued)
Contents of a .LST text file for a structured LIST:
/ comment lines starting by `/'
<chapter_1><edit_field_1>
<chapter_2><edit_field_2>
...
END
CASE_START=<control_character> default: first non-empty line
CASE_END=<control_character> default: first empty line
LIST FIELDS:
1 <type> <length> <field name> <identification key>
2 <type> <length> <field name> <identification key>
...
END
<comment lines>
To a great extent the definition of fields starting from a LIST FIELDS:
line is the same as in the FILE CREATE operation (see FCREATE?).
The essential addition is the <definition> part which is needed
for the identification of the field within the text.
Structured LISTs (continued)
Definition of fields:
As an example, one field in LIST FIELDS section can be given in the form
3 S 16 Name1 SHADOW:1,STRING
where `3' is # of the field `S' the type (string), `16' the length,
and `Name1' the name of the field.
The <identification key> is in this case `SHADOW:1,STRING' which means
that the first red string (i.e. string with consecutive shadow characters 1)
in the current case will be taken as the value of field `Name1'
Thus, if the first line in the current case with 1's as shadow characters is
This is the SURVO 84C System.
the field `Name1' will have the value `SURVO 84C'.
In the list of various identification keys which follows on next pages
this alternative is described briefly as
SHADOW:1,STRING identification key
This is the SURVO 84C System. essential part of the case text
SURVO 84C value obtained by identification
Identification keys in structured LISTs
Keys that are already implemented are indicated by (*).
Using shadow characters:
SHADOW:1,STRING (*) identification key
This is the SURVO 84C System. essential part of the case text
SURVO 84C value obtained by identification
SHADOW:1,WORD (*)
SURVO 84C System is
System
SHADOW:( - SHADOW:) (*)
This is the SURVO 84C System.
( ) shadow characters
SURVO 84C
SHADOW:17 - SHADOW:71 (*)
A field (this is it) is indicated by pairs (triples etc.) of shadows.
this is it
Identification keys in structured LISTs
Using shadow characters: (continued)
#SHADOW:1,STRING (*)
#SHADOW:1,WORD (*)
number of strings or words indicated by shadow `1'
Identification keys in structured LISTs
Using the control column:
*CONTROL:+ (*)
+ the entire line
* the entire line
CONTROL:[ - CONTROL:] (*)
defines a subcase starting from a line with `[' in the control column and
ending to a line with `]' in the control column (the latter line also included).
A subcase is always defined as a field with type and length `-' and
any field within it is defined as follows:
5 - - PART1 CONTROL:{ - CONTROL:}
6 S 16 KEYW SHADOW:1,WORD [PART1]
`PART1' in brackets in the definition of `KEYW' tells that the field
to be identified is the first word within the subcase `PART1' indicated
by the shadow character 1.
Identification keys in structured LISTs
Using the control column: (continued)
*CONTROL:+,12 *CONTROL:+,WORD,2
+01234567890123456 +SURVO 84C
*012345678901 *84C
*CONTROL:+,4,5 *CONTROL:+,WORD,2,END
+01234567890123456 +SURVO 84C System
*34567 *84C System
*CONTROL:+,WORD
+SURVO 84C
*SURVO
Identification keys in structured LISTs
Using keywords within the text itself:
TEXT:Birth_place:_
(any text) Birth place: Helsinki
Helsinki
TEXT:Birth_place:_,16 (*)
(any text) Birth place: Helsinki 1234567890
Helsinki 1234567
TEXT:Birth_place:_,WORD
(any text) Birth place: Helsinki 1234567890
Helsinki
Identification keys in structured LISTs
Miscellaneous:
WORD#:3 [S] (*)
the third word (in the subcase S).
#WORDS: [S] (*)
number of words (in subcase S).
#WORDS:word1,word2,... [S] (*)
number of words word1,word2,... (in subcase S).
Also wild card notations of the form *word,word*,*word* permitted.
#WORDS:#<keyword>/<text_file>] [S] (*)
as above with a list of words in <text_file> in the form
#<keyword> word1,word2,....
wordn,#END
In the two last items comparisons are case-insensitive by default.
To make them case-sensitive, insert `C' as follows:
#WORDS:C:word1,word2,... [S] (*)
Identification keys in structured LISTs
#LINES: [S] (*)
START:n [S]
n first bytes (control column excluded)
L = More information on LIST operations
Help System (web edition)