Myanmar: simplify reordering logic.

opentype-shaping-myanmar.md

@@ -340,10 +340,10 @@ Processing a run of `<mym2>` text involves five top-level stages:
 
 
 As with other Brahmi-derived and Indic scripts, the initial reordering
-stage and the final reordering stage each involve applying a set of several
-script-specific rules. The basic substitution features must be applied
-to the run in a specific order. The remaining substitution features in
-stage four, however, do not have a mandatory order.
+stage involves applying a set of several script-specific rules. The
+basic substitution features must be applied to the run in a specific
+order. The remaining substitution features in stage four, however, do
+not have a mandatory order. 
 
 
 Myanmar exhibits many of the same shaping patterns found in Indic
@@ -738,37 +738,28 @@ The algorithm for determining the base consonant is
         next consonant. 
   - The consonant stopped at will be the base consonant.
 
-> Note: The algorithm considers only `CONSONANT` class consonants, 
+<!--- > Note: The algorithm considers only `CONSONANT` class consonants, --->
 
 
-#### Stage 2, step 2: Tag matras ####
+#### Stage 2, step 2: Kinzi ####
 
-Second, all left-side dependent-vowel (matra) signs must be tagged to be
-moved to the beginning of the syllable, with `POS_PREBASE_MATRA`.
-
-All right-side and above-base dependent-vowel (matra)
-signs are tagged `POS_AFTER_SUBJOINED`.
-
-All below-base dependent-vowel (matra) signs are tagged
-`POS_BELOWBASE_CONSONANT`. 
+Second, initial <samp>"Kinzi"</samp>-triggering sequences that will become <samp>"Kinzi"</samp>s
+must be tagged with `POS_AFTER_MAIN`.
 
-For simplicity, shaping engines may choose to tag matras
-in an earlier text-processing step, using the information in the
-_Mark-placement subclass_ column of the character tables. It is
-critical at this step, however, that all matras correctly tagged
-before proceeding to the next step. 
+The sequences are:
 
-#### Stage 2, step 3: Anusvara ####
+  - <samp>"Ra,Asat,Halant"</samp>
+  - <samp>"Nga,Asat,Halant"</samp>
+  - <samp>"Mon Nga,Asat,Halant"</samp>
 
-Third, any `ANUSVARA` marks appearing immediately after a below-base
-vowel sign must be tagged with `POS_BEFORE_SUBJOINED`, so that the
-marks are reordered to a position immediately before the below-base
-vowel signs.
+In the Myanmar (or Burmese) language, <samp>"Nga"</samp> is the only <samp>"Kinzi"</samp>-forming
+consonant. <samp>"Mon Nga"</samp> can form a <samp>"Kinzi"</samp> in the Mon language, and <samp>"Ra"</samp>
+can form a <samp>"Kinzi"</samp> in Sanskrit written with the Myanmar script.
 
 
-#### Stage 2, step 4: Pre-base-reordering consonants ####
+#### Stage 2, step 3: Pre-base-reordering consonants ####
 
-Fourth, all pre-base-reordering consonants must be tagged with
+Third, all pre-base-reordering consonants must be tagged with
 `POS_PREBASE_CONSONANT`. 
 
 Myanmar has one pre-base-reordering consonant: <samp>"Medial Ra"</samp>.
@@ -780,54 +771,46 @@ Pre-base-reordering Medial Ra
 :::
 
 
-#### Stage 2, step 5: Kinzi ####
+#### Stage 2, step 4: Pre-base matras ####
 
-Fifth, initial <samp>"Kinzi"</samp>-triggering sequences that will become <samp>"Kinzi"</samp>s
-must be tagged with `POS_AFTER_MAIN`.
-
-The sequences are:
-
-  - <samp>"Ra,Asat,Halant"</samp>
-  - <samp>"Nga,Asat,Halant"</samp>
-  - <samp>"Mon Nga,Asat,Halant"</samp>
+Fourth, all left-side dependent-vowel (matra) signs must be tagged to be
+moved to the beginning of the syllable, with `POS_PREBASE_MATRA`.
 
-In the Myanmar (or Burmese) language, <samp>"Nga"</samp> is the only <samp>"Kinzi"</samp>-forming
-consonant. <samp>"Mon Nga"</samp> can form a <samp>"Kinzi"</samp> in the Mon language, and <samp>"Ra"</samp>
-can form a <samp>"Kinzi"</samp> in Sanskrit written with the Myanmar script.
+For simplicity, shaping engines may choose to tag matras
+in an earlier text-processing step, using the information in the
+_Mark-placement subclass_ column of the character tables. It is
+critical at this step, however, that all matras correctly tagged
+before proceeding to the next step. 
 
 
-#### Stage 2, step 6: Post-base consonants ####
+#### Stage 2, step 5: Syllables without below-base matras ####
 
-Sixth, any remaining non-base consonants that occur after the base
-consonant must be tagged with `POS_AFTER_MAIN`. Full consonants (of
-class `CONSONANT`) will be preceded by a <samp>"Halant"</samp> glyph. Medial
-consonants (of class `CONSONANT_MEDIAL`) will not be preceded by a
-<samp>"Halant"</samp> glyph. 
+Fifth, if the syllable contains no below-base dependent-vowel (matra)
+signs, then all of the remaining codepoints can be tagged with
+`POS_AFTER_MAIN`.
 
-> Note: <samp>"Medial Ra"</samp> should have been tagged with
-> `POS_PREBASE_CONSONANT` in stage 2, step four, and must not be
-> re-tagged in this step.
 
+#### Stage 2, step 6: Syllables with below-base-matras ####
 
+Sixth, if the syllable contains any below-base dependent-vowel (matra)
+signs, then those below-base matra signs must be tagged with
+`POS_BELOWBASE_CONSONANT`.
 
-#### Stage 2, step 7: Mark tagging ####
+All of the codepoints that precede the below-base dependent-vowel
+signs, but which were not already tagged in steps 1 through 4, must
+now be tagged with `POS_AFTER_MAIN`.
 
-<!--- not sure this is done!!! --->
+Any `ANUSVARA` marks that appear after the below-base dependent vowel
+signs must be tagged wtih `POS_BEFORE_SUBJOINED`.
 
-Seventh, all marks must be tagged with the same positioning tag as the
-closest non-mark character the mark has affinity with, so that they move together
-during the sorting step.
+All remaining codepoints that appear after the below-base
+dependent-vowel signs can be tagged with `POS_AFTER_SUBJOINED`.
 
-For all marks preceding the base consonant, the mark must be tagged
-with the same positioning tag as the closest preceding non-mark
-consonant.
 
-For all marks occurring after the base consonant, the mark must be
-tagged with the same positioning tag as the closest subsequent consonant.
+#### Stage 2, step 7: Variation selectors ####
 
-> Note: In this step, joiner and non-joiner characters must also be
-> tagged according to the same rules given for marks, even though
-> these characters are not categorized as marks in Unicode.
+Seventh, all Variation Selector codepoints must be tagged with the
+same positioning tag as the immediately preceding character.
 
 
 With these steps completed, the syllable can be sorted into the final sort order.