| 1 |
<?xml version="1.0" encoding="utf-8"?> |
| 2 |
<!DOCTYPE chapter SYSTEM "docbook-dtd-45/docbookx.dtd"> |
| 3 |
|
| 4 |
<chapter id="subreadings"> |
| 5 |
<title>Sub-Readings</title> |
| 6 |
|
| 7 |
<para> |
| 8 |
Sub-readings introduce a bit of hierarchy into readings, letting a reading have a hidden reading attached to it, |
| 9 |
which in turn may have another hidden reading, and so on. See the <code>test/T_SubReading_Apertium</code> and |
| 10 |
<code>test/T_SubReading_CG</code> tests for usage examples. |
| 11 |
</para> |
| 12 |
|
| 13 |
<section id="sub-stream-apertium"> |
| 14 |
<title>Apertium Format</title> |
| 15 |
|
| 16 |
<para> |
| 17 |
The Apertium stream format supports sub-readings via the + delimiter for readings. E.g. |
| 18 |
</para> |
| 19 |
<screen> |
| 20 |
^word/aux3<tag>+aux2<tag>+aux1<tag>+main<tag>$ |
| 21 |
</screen> |
| 22 |
<para> |
| 23 |
is a cohort with 1 reading which has a three level deep sub-reading. The order of which is the primary reading |
| 24 |
vs. sub-readings depends on the grammar SUBREADINGS setting: |
| 25 |
</para> |
| 26 |
<screen> |
| 27 |
SUBREADINGS = RTL ; # Default, right-to-left |
| 28 |
SUBREADINGS = LTR ; # Alternate, left-to-right |
| 29 |
</screen> |
| 30 |
<para> |
| 31 |
In default RTL mode, the above reading has the primary reading <code>"main"</code> with sub-reading |
| 32 |
<code>"aux1"</code> with sub-reading <code>"aux2"</code> and finally sub-reading <code>"aux3"</code>. |
| 33 |
</para> |
| 34 |
<para> |
| 35 |
In LTR mode, the above reading has the primary reading <code>"aux3"</code> with sub-reading |
| 36 |
<code>"aux2"</code> with sub-reading <code>"aux1"</code> and finally sub-reading <code>"main"</code>. |
| 37 |
</para> |
| 38 |
</section> |
| 39 |
|
| 40 |
<section id="sub-stream-cg"> |
| 41 |
<title>CG Format</title> |
| 42 |
|
| 43 |
<para> |
| 44 |
The CG stream format supports sub-readings via indentation level. E.g. |
| 45 |
</para> |
| 46 |
<screen> |
| 47 |
"<word>" |
| 48 |
"main" tag |
| 49 |
"aux1" tag |
| 50 |
"aux2" tag |
| 51 |
"aux3" tag |
| 52 |
</screen> |
| 53 |
<para> |
| 54 |
is a cohort with 1 reading which has a three level deep sub-reading. Unlike the Apertium format, the order is |
| 55 |
strictly defined by indentation and cannot be changed. The above reading has the primary reading |
| 56 |
<code>"main"</code> with sub-reading <code>"aux1"</code> with sub-reading <code>"aux2"</code> |
| 57 |
and finally sub-reading <code>"aux3"</code>. |
| 58 |
</para> |
| 59 |
<para> |
| 60 |
The indentation level is detected on a per-cohort basis. All whitespace counts the same for purpose of determining |
| 61 |
indentation, so 1 tab is same as 1 space is same as 1 no-break space and so on. Since it is per-cohort, it won't |
| 62 |
matter if previous cohorts has a different indentation style, so it is safe to mix cohorts from multiple sources. |
| 63 |
</para> |
| 64 |
</section> |
| 65 |
|
| 66 |
<section id="sub-grammar"> |
| 67 |
<title>Grammar Syntax</title> |
| 68 |
|
| 69 |
<para> |
| 70 |
Working with sub-readings involves 2 new grammar features: Rule Option SUB:N and Contextual Option /N. |
| 71 |
</para> |
| 72 |
|
| 73 |
<section id="sub-grammar-subn"> |
| 74 |
<title>Rule Option SUB:N</title> |
| 75 |
<indexterm> |
| 76 |
<primary>SUB:N</primary> |
| 77 |
</indexterm> |
| 78 |
|
| 79 |
<para> |
| 80 |
Rule option SUB:N tells a rule which sub-reading it should operate on and which it should test as target. |
| 81 |
The N is an integer in the range -2^31 to 2^31. SUB:0 is the primary reading and same as not specifying SUB. |
| 82 |
Positive numbers refer to sub-readings starting from the primary and going deeper, while negative numbers start |
| 83 |
from the last sub-reading and go towards the primary. Thus, SUB:-1 always refers to the deepest sub-reading. |
| 84 |
</para> |
| 85 |
<para> |
| 86 |
Given the above CG input and the rules |
| 87 |
<screen> |
| 88 |
ADD SUB:-1 (mark) (*) ; |
| 89 |
ADD SUB:1 (twain) (*) ; |
| 90 |
</screen> |
| 91 |
the output will be |
| 92 |
<screen> |
| 93 |
"<word>" |
| 94 |
"main" tag |
| 95 |
"aux1" tag twain |
| 96 |
"aux2" tag |
| 97 |
"aux3" tag mark |
| 98 |
</screen> |
| 99 |
Note that SUB:N also determines which reading is looked at as target, so it will work for all rule types. |
| 100 |
</para> |
| 101 |
</section> |
| 102 |
|
| 103 |
<section id="sub-grammar-slashn"> |
| 104 |
<title>Contextual Option /N</title> |
| 105 |
<indexterm> |
| 106 |
<primary>/N</primary> |
| 107 |
<secondary>/*</secondary> |
| 108 |
</indexterm> |
| 109 |
|
| 110 |
<para> |
| 111 |
Context option /N tests the N'th sub-reading of the currently active reading, where N follows the same rules |
| 112 |
as for SUB:N above. The /N must be last in the context position. |
| 113 |
</para> |
| 114 |
<para> |
| 115 |
If N is * then the test will search the main reading and all sub-readings. |
| 116 |
</para> |
| 117 |
<para> |
| 118 |
Given the above CG input and the rules |
| 119 |
<screen> |
| 120 |
ADD (mark) (*) (0/-1 ("aux3")) ; # matches 3rd sub-reading "aux3" |
| 121 |
ADD (twain) (*) (0/1 ("aux1")) ; # matches 1st sub-reading "aux1" |
| 122 |
ADD (writes) (*) (0/1 ("main")) ; # won't match as 1st sub-reading doesn't have tag "main" |
| 123 |
</screen> |
| 124 |
the output will be |
| 125 |
<screen> |
| 126 |
"<word>" |
| 127 |
"main" tag mark twain |
| 128 |
"aux1" tag |
| 129 |
"aux2" tag |
| 130 |
"aux3" tag |
| 131 |
</screen> |
| 132 |
</para> |
| 133 |
</section> |
| 134 |
</section> |
| 135 |
|
| 136 |
</chapter> |