From 615b3ede627355d5cbbde38f3d11326dee23f14a Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Tue, 5 May 2026 19:41:19 +0700 Subject: [PATCH 01/42] Added Pali. --- _config.yml | 2 ++ _pi/template-index.md | 59 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 61 insertions(+) create mode 100644 _pi/template-index.md diff --git a/_config.yml b/_config.yml index 8c63aa8d685..b6b79980639 100644 --- a/_config.yml +++ b/_config.yml @@ -539,6 +539,8 @@ collections: output: true fsl: output: true + pi: + output: true # End of language collections (do not remove this comment, addlanguage.sh looks for it) diff --git a/_pi/template-index.md b/_pi/template-index.md new file mode 100644 index 00000000000..197b00555d4 --- /dev/null +++ b/_pi/template-index.md @@ -0,0 +1,59 @@ +--- +layout: base +title: 'Pali UD' +udver: '2' +--- + +# UD for Pali + +## Tokenization and Word Segmentation + +* + +--- +**Instruction**: Describe the general rules for delimiting words (for example, based on whitespace and punctuation) and exceptions to these rules. Specify whether words with spaces and/or multiword tokens occur. Include links to further language-specific documentation if available. + +--- + +## Morphology + +### Tags + +* + +--- +**Instruction**: Specify any unused tags. Explain what words are tagged as PART. Describe how the AUX-VERB and DET-PRON distinctions are drawn, and specify whether there are (de)verbal forms tagged as ADJ, ADV or NOUN. Include links to language-specific tag definitions if any. + +--- + +### Features + +* + +--- +**Instruction**: Describe inherent and inflectional features for major word classes (at least NOUN and VERB). Describe other noteworthy features. Include links to language-specific feature definitions if any. + +--- + +## Syntax + +* + +--- +**Instruction**: Give criteria for identifying core arguments (subjects and objects), and describe the range of copula constructions in nonverbal clauses. List all subtype relations used. Include links to language-specific relations definitions if any. + +--- + +## Treebanks + +There are [N](../treebanks/pi-comparison.html) Pali UD treebanks: + + * [Pali-A](../treebanks/pi_a/index.html) + * [Pali-B](../treebanks/pi_b/index.html) + +--- +**Instruction**: Treebank-specific pages are generated automatically from the README file in the treebank repository and +from the data in the latest release. Link to the respective `*-index.html` page in the `treebanks` folder, using the language code +and the treebank code in the file name. + +--- From 49008297d9ba1e0c5be226a9da2bc1f4e2120e12 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Tue, 5 May 2026 20:13:30 +0700 Subject: [PATCH 02/42] Add orthography and tokenization sections --- _pi/template-index.md | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/_pi/template-index.md b/_pi/template-index.md index 197b00555d4..ecf48a42c1c 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -6,14 +6,18 @@ udver: '2' # UD for Pali -## Tokenization and Word Segmentation +## Orthography -* +* While Pāli can traditionally be written with a variety of scripts, for UD we use [the IAST Romanization standard](https://en.wikipedia.org/wiki/International_Alphabet_of_Sanskrit_Transliteration). +* If your corpus or text is in another script or alternate Romanization, you can use a tool like [aksharamukha](https://aksharamukha.com/python) to automatically convert it to IAST. +* As different scripts use different, and sometimes no, punctuation, we further standardize our orthography by **removing all punctuation** from our texts. +* IAST is case insensitive. Here, we use lowercase with the first letter each sentence capitalized by convention. ---- -**Instruction**: Describe the general rules for delimiting words (for example, based on whitespace and punctuation) and exceptions to these rules. Specify whether words with spaces and/or multiword tokens occur. Include links to further language-specific documentation if available. +## Tokenization and Word Segmentation ---- +* Words are generally delimited by spaces. +* Compounds are split and sandhi undone as it is the custom in Sanskritic linguistics. +* Multi-word tokens are generally not used. Parts of a multi-word name (e.g. "Saccaka Nigaṇṭhāputta") should get the [flat:name]() dependency relation. ## Morphology From 75c0b2f4768ed0ea139ca7e04933c72c3f7d82e7 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Wed, 6 May 2026 19:11:51 +0700 Subject: [PATCH 03/42] Add root compound documentation --- _pi/dep/compound.md | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) create mode 100644 _pi/dep/compound.md diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md new file mode 100644 index 00000000000..24e85068403 --- /dev/null +++ b/_pi/dep/compound.md @@ -0,0 +1,25 @@ +--- +layout: relation +title: 'compound' +shortdef: 'compound' +udver: '2' +--- + +The `compound` relation is used for Pāli compounds to show how their parts relate to each other. + +A Pāli compound is identified through the fact that only the final member of the compound is grammatically inflected, and the earlier part(s) of the compound are in their uninflected form(s). +For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form and only "fallen" *patito* is declined. + +Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is in traditional (Pāṇinian) analysis: + + * [compound:av]() — *avyayībhāva* (an adverbial compound) + * [compound:dg]() — *dvigu* (a numerical relation) + * [compound:dv]() — *dvandva* (a coordinative relation) + * [compound:kd]() — *karmadhāraya* (an adjectival relation) + * [compound:tp]() — *tatpuruṣa* (a case relation) + * [compound:na]() - *nañtatpuruṣa* (a negative relation) + +Note that we don't have a *bahuvrīhi* type, because *bahuvrīhi* compounds are internally some other type and are only *bahuvrīhi* by virtue of how they are used in the sentence. Since these `compound` dependency relations are strictly internal to the compound, we don't need a `compound:bv` relation and can use e.g. [amod]() on the final token instead to indicate the e.g. adjectival relationship of the compound as a whole. + +Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. + From 714d143f4c59a4158d30dd8b202a4316e924eaaa Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Thu, 7 May 2026 08:24:16 +0700 Subject: [PATCH 04/42] Add 'ti info --- _pi/dep/compound.md | 2 +- _pi/template-index.md | 9 ++++++++- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md index 24e85068403..4706d8b0a52 100644 --- a/_pi/dep/compound.md +++ b/_pi/dep/compound.md @@ -19,7 +19,7 @@ Because the last part of the compound is inflected, we treat that part as the ro * [compound:tp]() — *tatpuruṣa* (a case relation) * [compound:na]() - *nañtatpuruṣa* (a negative relation) -Note that we don't have a *bahuvrīhi* type, because *bahuvrīhi* compounds are internally some other type and are only *bahuvrīhi* by virtue of how they are used in the sentence. Since these `compound` dependency relations are strictly internal to the compound, we don't need a `compound:bv` relation and can use e.g. [amod]() on the final token instead to indicate the e.g. adjectival relationship of the compound as a whole. +Note that we don't have a *bahuvrīhi* type, because *bahuvrīhi* compounds are internally some other type and are only *bahuvrīhi* by virtue of how they are used in the sentence. Since these `compound` dependency relations are strictly internal to the compound, we don't need a `compound:bv` relation and can use e.g. [amod]() on the final token instead to indicate the e.g. adjectival nature of the compound as a whole. Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. diff --git a/_pi/template-index.md b/_pi/template-index.md index ecf48a42c1c..08a5828eece 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -23,7 +23,14 @@ udver: '2' ### Tags -* +#### Particles + +The following particles should get the `UPOS` `PART`: + +* The quote-ending particle *ti* + * `LEMMA`: *iti* + * `DEPREL`: `mark` to the head of the quoted clause + * The preceding token should usually get `SpaceAfter=No` (as appropriate) --- **Instruction**: Specify any unused tags. Explain what words are tagged as PART. Describe how the AUX-VERB and DET-PRON distinctions are drawn, and specify whether there are (de)verbal forms tagged as ADJ, ADV or NOUN. Include links to language-specific tag definitions if any. From 6291117a78b86a4d34d68991f4dbfea94ac9a69a Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Thu, 7 May 2026 08:33:37 +0700 Subject: [PATCH 05/42] participles --- _pi/template-index.md | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/_pi/template-index.md b/_pi/template-index.md index 08a5828eece..0a77abbf377 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -23,6 +23,16 @@ udver: '2' ### Tags +#### Participles + +Participles are tagged as `VERB`s with `FEATS` as follows: +* `VerbForm=Part` +* `Voice=Act` or `Voice=Pass` +* Full nominal agreement (Case, Gender, Number) + +The participle's `LEMMA` should be the 3rd-person singular present indicative of the underlying verb +(e.g. *pareta* "afflicted" has the lemma *pareti* "goes beyond"). + #### Particles The following particles should get the `UPOS` `PART`: From d577cf9727ceba359461d5636c254b6e963217b2 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Thu, 7 May 2026 08:48:19 +0700 Subject: [PATCH 06/42] verb features --- _pi/template-index.md | 38 +++++++++++++++++++++++++++----------- 1 file changed, 27 insertions(+), 11 deletions(-) diff --git a/_pi/template-index.md b/_pi/template-index.md index 0a77abbf377..8fb6503bb91 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -23,19 +23,11 @@ udver: '2' ### Tags -#### Participles - -Participles are tagged as `VERB`s with `FEATS` as follows: -* `VerbForm=Part` -* `Voice=Act` or `Voice=Pass` -* Full nominal agreement (Case, Gender, Number) - -The participle's `LEMMA` should be the 3rd-person singular present indicative of the underlying verb -(e.g. *pareta* "afflicted" has the lemma *pareti* "goes beyond"). +Participles are tagged as `VERB`s and not `ADJ`s or `ADV`s. #### Particles -The following particles should get the `UPOS` `PART`: +The following particles get the `PART` `UPOS`: * The quote-ending particle *ti* * `LEMMA`: *iti* @@ -49,7 +41,31 @@ The following particles should get the `UPOS` `PART`: ### Features -* +#### Verbs + +All verb forms in Pāli should get the `VERB` `UPOS` +and their `LEMMA` is the 3rd-person singular present indicative of the underlying verb. + +For example, the past participle *gata* "gone" has the `LEMMA` *gacchati* "goes". + +##### Finite Verbs + +Should have the `VerbForm=Fin` along with `Mood`, `Tense`, `Person`, `Number`, and `Voice` as appropriate. + +##### Absolutives + +Absolutives formed in Pāli with *-tvā* / *-ya* get the `FEATS` `VerbForm=Conv` and are attached to the main verb via `advcl`. + +##### Infinitives + +Infinitives in Pāli end in *-tuṃ*, have the `VerbForm=Inf` and are typically `xcomp`. + +##### Participles + +Participles have the following `FEATS`: +* `VerbForm=Part` +* `Voice=Act` or `Voice=Pass` +* Full nominal agreement (Case, Gender, Number) --- **Instruction**: Describe inherent and inflectional features for major word classes (at least NOUN and VERB). Describe other noteworthy features. Include links to language-specific feature definitions if any. From 458d09a08f8e73240e143fde24ff31d1e46fd253 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Thu, 7 May 2026 10:09:57 +0700 Subject: [PATCH 07/42] rest of prompt doc --- _pi/template-index.md | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/_pi/template-index.md b/_pi/template-index.md index 8fb6503bb91..e25ed7a0cb0 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -6,6 +6,8 @@ udver: '2' # UD for Pali +Pāli is an Indo-Aryan language, widely studied as the liturgical language of Theravāda Buddhism. + ## Orthography * While Pāli can traditionally be written with a variety of scripts, for UD we use [the IAST Romanization standard](https://en.wikipedia.org/wiki/International_Alphabet_of_Sanskrit_Transliteration). @@ -13,11 +15,17 @@ udver: '2' * As different scripts use different, and sometimes no, punctuation, we further standardize our orthography by **removing all punctuation** from our texts. * IAST is case insensitive. Here, we use lowercase with the first letter each sentence capitalized by convention. +## Sentence Segmentation + +While we strip out punctuation in our `# text` rows, we do recommend using any punctuation in your source to help segment into sentences. In the Pāli Text Society and Mahāsaṅgīti editions, you should generally use both semicolons (`;`) and periods (`.`) to split sentences. However, be aware that Pāli verses may sometimes use semicolons to mark *pāda* "line" boundaries which are not semantic. Always segment sentences by syntax. + ## Tokenization and Word Segmentation * Words are generally delimited by spaces. * Compounds are split and sandhi undone as it is the custom in Sanskritic linguistics. -* Multi-word tokens are generally not used. Parts of a multi-word name (e.g. "Saccaka Nigaṇṭhāputta") should get the [flat:name]() dependency relation. +* Compounds receive token ranges (e.g. `10-12`) and should get the appropriate [compound]() relation subtype. +* Sandhi should be marked with a `SpaceAfter=No` in the earlier token's `MISC`. +* Multi-word tokens are generally not used. Parts of a multi-word name (e.g. "Saccaka Nigaṇṭhāputta") should get the [flat:name]() dependency relation. ## Morphology @@ -33,6 +41,10 @@ The following particles get the `PART` `UPOS`: * `LEMMA`: *iti* * `DEPREL`: `mark` to the head of the quoted clause * The preceding token should usually get `SpaceAfter=No` (as appropriate) +* The negation particle *na* (and its sandhi variants) + * The lemma is *na* + * Mark it as `Polarity=Neg` in `FEATS` + * Give it the [advmod]() relation to the negated head. --- **Instruction**: Specify any unused tags. Explain what words are tagged as PART. Describe how the AUX-VERB and DET-PRON distinctions are drawn, and specify whether there are (de)verbal forms tagged as ADJ, ADV or NOUN. Include links to language-specific tag definitions if any. From 12249995b8e759a2bbf8395b556a9eed07186959 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Thu, 7 May 2026 10:28:21 +0700 Subject: [PATCH 08/42] ambiguous features --- _pi/template-index.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/_pi/template-index.md b/_pi/template-index.md index e25ed7a0cb0..cbb1c9ecb86 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -53,6 +53,8 @@ The following particles get the `PART` `UPOS`: ### Features +Some forms in Pāli are ambiguous. For example, a feminine noun of the `-ā` class with a `-āya` suffix might be Singular Instrumental, Ablative, Genitive, Dative, or Locative! When `Case` can be inferred from context, feel free to mark just the semantically correct case. If multiple parses are reasonable in a given context, list all the semantically plausible values in alphabetical order, separated by commas (e.g. if you have an *-āya* noun, and all but instrumental are reasonable parses, mark it `Case=Abl,Dat,Gen,Loc|Gender=Fem|Number=Sing`). + #### Verbs All verb forms in Pāli should get the `VERB` `UPOS` From d771baf4063de8b891cdcac3a9a0983cadd323c6 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Thu, 7 May 2026 10:59:50 +0700 Subject: [PATCH 09/42] nits --- _pi/dep/compound.md | 4 ++-- _pi/template-index.md | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md index 4706d8b0a52..386004bd3b9 100644 --- a/_pi/dep/compound.md +++ b/_pi/dep/compound.md @@ -5,10 +5,10 @@ shortdef: 'compound' udver: '2' --- -The `compound` relation is used for Pāli compounds to show how their parts relate to each other. +The `compound` relation subtypes are used for Pāli compounds to show how the compound's parts relate to each other. A Pāli compound is identified through the fact that only the final member of the compound is grammatically inflected, and the earlier part(s) of the compound are in their uninflected form(s). -For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form and only "fallen" *patito* is declined. +For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form and only "fallen" *patito* declined. Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is in traditional (Pāṇinian) analysis: diff --git a/_pi/template-index.md b/_pi/template-index.md index cbb1c9ecb86..230d787f8d0 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -13,7 +13,7 @@ Pāli is an Indo-Aryan language, widely studied as the liturgical language of Th * While Pāli can traditionally be written with a variety of scripts, for UD we use [the IAST Romanization standard](https://en.wikipedia.org/wiki/International_Alphabet_of_Sanskrit_Transliteration). * If your corpus or text is in another script or alternate Romanization, you can use a tool like [aksharamukha](https://aksharamukha.com/python) to automatically convert it to IAST. * As different scripts use different, and sometimes no, punctuation, we further standardize our orthography by **removing all punctuation** from our texts. -* IAST is case insensitive. Here, we use lowercase with the first letter each sentence capitalized by convention. +* IAST is case insensitive. Here, we use lowercase, with the first letter of each sentence capitalized by convention. ## Sentence Segmentation @@ -64,7 +64,7 @@ For example, the past participle *gata* "gone" has the `LEMMA` *gacchati* "goes" ##### Finite Verbs -Should have the `VerbForm=Fin` along with `Mood`, `Tense`, `Person`, `Number`, and `Voice` as appropriate. +Should have the `VerbForm=Fin` along with `Mood`, `Tense`, `Person`, `Number`, and `Voice`. ##### Absolutives @@ -79,7 +79,7 @@ Infinitives in Pāli end in *-tuṃ*, have the `VerbForm=Inf` and are typically Participles have the following `FEATS`: * `VerbForm=Part` * `Voice=Act` or `Voice=Pass` -* Full nominal agreement (Case, Gender, Number) +* Full nominal agreement features (Case, Gender, Number) --- **Instruction**: Describe inherent and inflectional features for major word classes (at least NOUN and VERB). Describe other noteworthy features. Include links to language-specific feature definitions if any. From 0107766cac20ea9c464c304337d05136e21dca68 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Sat, 9 May 2026 11:47:26 +0700 Subject: [PATCH 10/42] A few updates from our convo online --- _pi/dep/compound.md | 5 ++--- _pi/template-index.md | 12 ++++++++++-- 2 files changed, 12 insertions(+), 5 deletions(-) diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md index 386004bd3b9..0fa86bb496d 100644 --- a/_pi/dep/compound.md +++ b/_pi/dep/compound.md @@ -12,14 +12,13 @@ For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form a Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is in traditional (Pāṇinian) analysis: - * [compound:av]() — *avyayībhāva* (an adverbial compound) * [compound:dg]() — *dvigu* (a numerical relation) * [compound:dv]() — *dvandva* (a coordinative relation) * [compound:kd]() — *karmadhāraya* (an adjectival relation) * [compound:tp]() — *tatpuruṣa* (a case relation) * [compound:na]() - *nañtatpuruṣa* (a negative relation) - -Note that we don't have a *bahuvrīhi* type, because *bahuvrīhi* compounds are internally some other type and are only *bahuvrīhi* by virtue of how they are used in the sentence. Since these `compound` dependency relations are strictly internal to the compound, we don't need a `compound:bv` relation and can use e.g. [amod]() on the final token instead to indicate the e.g. adjectival nature of the compound as a whole. + * [compound:bv]() - *bahuvrīhi* (a compound used adjectively, e.g. *sa-*) + * [compound:av]() — *avyayībhāva* (a compound used adverbially, e.g. *anu-*) Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. diff --git a/_pi/template-index.md b/_pi/template-index.md index 230d787f8d0..e1a917c8af2 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -24,7 +24,7 @@ While we strip out punctuation in our `# text` rows, we do recommend using any p * Words are generally delimited by spaces. * Compounds are split and sandhi undone as it is the custom in Sanskritic linguistics. * Compounds receive token ranges (e.g. `10-12`) and should get the appropriate [compound]() relation subtype. -* Sandhi should be marked with a `SpaceAfter=No` in the earlier token's `MISC`. +* Sandhi should be marked with a `SpaceAfter=No` in the earlier token's `MISC` and should not get a range id. * Multi-word tokens are generally not used. Parts of a multi-word name (e.g. "Saccaka Nigaṇṭhāputta") should get the [flat:name]() dependency relation. ## Morphology @@ -41,10 +41,18 @@ The following particles get the `PART` `UPOS`: * `LEMMA`: *iti* * `DEPREL`: `mark` to the head of the quoted clause * The preceding token should usually get `SpaceAfter=No` (as appropriate) -* The negation particle *na* (and its sandhi variants) +* The negation particle *na* / *n'* * The lemma is *na* * Mark it as `Polarity=Neg` in `FEATS` * Give it the [advmod]() relation to the negated head. +* *api* / *pi* + * The lemma is *api* + * It should normally get the [discourse]() dependency relation when used rhetorically, but can get `advmod` if used conjunctively. +* *eva* and *kho* + * Normally get the `discourse` dependency, except in `fixed` phrases +* *kira* + * When used as "apparently" it should also get the `discourse` relation. + * Note that "*kira*" is not a particle but a `VERB` if it is the imperative 2nd sg or the reflexive aorist 1st sg of *kirati*. --- **Instruction**: Specify any unused tags. Explain what words are tagged as PART. Describe how the AUX-VERB and DET-PRON distinctions are drawn, and specify whether there are (de)verbal forms tagged as ADJ, ADV or NOUN. Include links to language-specific tag definitions if any. From da4c9cff3b90bb3aac18d1d64347e6da20f68e53 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 11 May 2026 19:30:46 +0700 Subject: [PATCH 11/42] Add a couple more decisions --- _pi/dep/compound.md | 4 ++++ _pi/template-index.md | 1 + 2 files changed, 5 insertions(+) diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md index 0fa86bb496d..ccc7aa44c49 100644 --- a/_pi/dep/compound.md +++ b/_pi/dep/compound.md @@ -22,3 +22,7 @@ Because the last part of the compound is inflected, we treat that part as the ro Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. +"Compounds" that are only compounds because they are *na-* negated should **not** be treated as compounds. +If the *na-* is in "n'" form (on a verb, as in "*nānubhava*" "doesn't accomplish"), that should be treated as sandhi (as above). +If the *na-* is in "a-" prefix (on a adjective/participle, as in "*amakkhito*" "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. +If, however, the negated word is a compound even without the negation, as in "*anindārosa*" "free of blaming and offense", then the *na* should be treated as a compound part and given the `compound:na` relation in that case. diff --git a/_pi/template-index.md b/_pi/template-index.md index e1a917c8af2..c70bbbc5850 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -32,6 +32,7 @@ While we strip out punctuation in our `# text` rows, we do recommend using any p ### Tags Participles are tagged as `VERB`s and not `ADJ`s or `ADV`s. +This means that they cannot get the `amod` dependency relation when used to modify a noun, so we use `acl` instead to attach it to its target noun. #### Particles From 2dac4aa3ef26b4b25bbd7f24e358782b90d02880 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Sun, 17 May 2026 17:35:02 +0700 Subject: [PATCH 12/42] noun lemmas --- _pi/template-index.md | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) diff --git a/_pi/template-index.md b/_pi/template-index.md index c70bbbc5850..2e5d830dcea 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -29,11 +29,32 @@ While we strip out punctuation in our `# text` rows, we do recommend using any p ## Morphology -### Tags +### Lemmatization + +The `LEMMA` for all verb forms is the 3rd-person singular present indicative of the underlying verb. +For example, the past participle *gata* "gone" gets the `LEMMA` *gacchati*. + +#### Noun Lemmas + +Lemmas are always lowercase, even for proper nouns. + +For personal pronouns, collapse the cases down to the nominative. For example, *amhākaṃ* gets *mayaṃ* and *te* gets *tvaṃ*. Keep first and second, singular and plural separate. + +For demonstrative pronouns, collapse the cases, gender, and number to the nominative, singular, neuter form: + +* Use *tad* for all *ta-* forms: *sā*, *sa*, *taṃ*, *tena*, etc. +* Use *yad* for all *ya-* forms: *yo*, *yā*, *yaṃ*, *yamhi*, etc. +* Use *idaṃ* for its various forms: *ayaṃ*, *imaṃ*, *imāni*, *imehi*, *anena*, etc. + +Substantives and adjectives should get the stem form as their lemma. For example, *itthiyo* has the lemma *itthi*, *bhikkhave* gets the lemma *bhikkhu*, and *rasso* the lemma *rassa*. + +### Part of Speech Tags Participles are tagged as `VERB`s and not `ADJ`s or `ADV`s. This means that they cannot get the `amod` dependency relation when used to modify a noun, so we use `acl` instead to attach it to its target noun. +Because Pāli is originally an oral language, we are stripping out symbols and punctuation, meaning we do not use the `SYM` or `PUNCT` tags. + #### Particles The following particles get the `PART` `UPOS`: @@ -66,10 +87,7 @@ Some forms in Pāli are ambiguous. For example, a feminine noun of the `-ā` cla #### Verbs -All verb forms in Pāli should get the `VERB` `UPOS` -and their `LEMMA` is the 3rd-person singular present indicative of the underlying verb. - -For example, the past participle *gata* "gone" has the `LEMMA` *gacchati* "goes". +All verb forms in Pāli should get the `VERB` `UPOS`. ##### Finite Verbs From c1a7006aa35fc707cf8cdf2e5098cf573af1fde9 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Sun, 24 May 2026 18:00:06 +0700 Subject: [PATCH 13/42] Add DET reminder --- _pi/template-index.md | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/_pi/template-index.md b/_pi/template-index.md index 2e5d830dcea..3a7ef8ed554 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -46,7 +46,7 @@ For demonstrative pronouns, collapse the cases, gender, and number to the nomina * Use *yad* for all *ya-* forms: *yo*, *yā*, *yaṃ*, *yamhi*, etc. * Use *idaṃ* for its various forms: *ayaṃ*, *imaṃ*, *imāni*, *imehi*, *anena*, etc. -Substantives and adjectives should get the stem form as their lemma. For example, *itthiyo* has the lemma *itthi*, *bhikkhave* gets the lemma *bhikkhu*, and *rasso* the lemma *rassa*. +Substantives and adjectives should get the stem form as their lemma. For example, *itthiyo* has the lemma `itthi`, *bhikkhave* gets the lemma `bhikkhu`, *rasso* the lemma `rassa`, *rāja* the lemma `rājan`, *bhagavato* `bhagavant`, etc. ### Part of Speech Tags @@ -55,6 +55,10 @@ This means that they cannot get the `amod` dependency relation when used to modi Because Pāli is originally an oral language, we are stripping out symbols and punctuation, meaning we do not use the `SYM` or `PUNCT` tags. +#### Determiners + +If a pronoun such as *imaṃ* is used as a determiner (i.e. it gets deprel `det` pointing to a substantive) then it should get the UPOS `DET` instead of `PRON`. + #### Particles The following particles get the `PART` `UPOS`: @@ -95,7 +99,7 @@ Should have the `VerbForm=Fin` along with `Mood`, `Tense`, `Person`, `Number`, a ##### Absolutives -Absolutives formed in Pāli with *-tvā* / *-ya* get the `FEATS` `VerbForm=Conv` and are attached to the main verb via `advcl`. +Absolutives (sometimes called "gerunds") formed in Pāli with *-tvā* / *-ya* get the `VerbForm=Conv` feature and are attached to the main verb via `advcl`. ##### Infinitives From a557ff4dfba9303ccc7a5ca73d9350c7aed51bb8 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:18:25 +0200 Subject: [PATCH 14/42] Create documentation for 'obl:goal' relation Added documentation for the 'obl:goal' relation, including its definition and usage example. --- _pi/dep/obl-goal.md | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) create mode 100644 _pi/dep/obl-goal.md diff --git a/_pi/dep/obl-goal.md b/_pi/dep/obl-goal.md new file mode 100644 index 00000000000..cc35c662dd0 --- /dev/null +++ b/_pi/dep/obl-goal.md @@ -0,0 +1,19 @@ +--- +layout: relation +title: 'obl:goal' +shortdef: 'oblique argument expressing the goal' +udver: '2' +--- + +Used for a noun expressing the goal or direction of a movement. + +We wish to go to your dwellings (RV 1.154.6a). +~~~ sdparse +tā́ vām vā́stūni uśmasi gámadhyai \n to-these of-you-two habitations we-wish to-go + +nmod(vā́stūni, vām) +det(vā́stūni, tā́) +obl:goal(gámadhyai, vā́stūni) +xcomp(uśmasi, gámadhyai) +~~~ + From 37111b94bb8dfeb0adf9e646993dc1e573f469e6 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:24:07 +0200 Subject: [PATCH 15/42] Revise content in obl-goal.md Updated example sentences and grammatical annotations in the document. --- _pi/dep/obl-goal.md | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/_pi/dep/obl-goal.md b/_pi/dep/obl-goal.md index cc35c662dd0..b15b70d4bb1 100644 --- a/_pi/dep/obl-goal.md +++ b/_pi/dep/obl-goal.md @@ -9,11 +9,11 @@ Used for a noun expressing the goal or direction of a movement. We wish to go to your dwellings (RV 1.154.6a). ~~~ sdparse -tā́ vām vā́stūni uśmasi gámadhyai \n to-these of-you-two habitations we-wish to-go +bodhisatto ... **mātukucchiṃ** okkami +bodhisattva-NOM.M.SG ... mother-womb-AKK.SG.M +'Bodisattva descended into the womb of his mother.' + +obl:goal(mātukucchiṃ) -nmod(vā́stūni, vām) -det(vā́stūni, tā́) -obl:goal(gámadhyai, vā́stūni) -xcomp(uśmasi, gámadhyai) ~~~ From 709fb2f4ee1dba040c7c6bb7be13822488df419c Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:29:57 +0200 Subject: [PATCH 16/42] Create documentation for 'obl:agent' relation Added documentation for the 'obl:agent' relation, including its definition and usage example. --- _pi/dep/obl-agent.md | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) create mode 100644 _pi/dep/obl-agent.md diff --git a/_pi/dep/obl-agent.md b/_pi/dep/obl-agent.md new file mode 100644 index 00000000000..27ff9457b4c --- /dev/null +++ b/_pi/dep/obl-agent.md @@ -0,0 +1,16 @@ +--- +layout: relation +title: 'obl:agent' +shortdef: 'oblique argument expressing the goal' +udver: '2' +--- + +Used for a noun expressing the agent of an action. + +~~~ sdparse +Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena +'And the mother of a bodhisattva will be violated by a man of lustful intent.' + +obl:agent (purisena) +~~~ + From 6a945e3e41d641e5ecf7443e04089d7dddbd8e56 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:44:29 +0200 Subject: [PATCH 17/42] Update compound-dg.md with compound examples Added examples of compound relations and their annotations. --- _pi/dep/compound-dg.md | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) create mode 100644 _pi/dep/compound-dg.md diff --git a/_pi/dep/compound-dg.md b/_pi/dep/compound-dg.md new file mode 100644 index 00000000000..e8881232227 --- /dev/null +++ b/_pi/dep/compound-dg.md @@ -0,0 +1,29 @@ +--- +layout: relation +title: 'compound' +shortdef: 'compound' +udver: '2' +--- + +The `compound` relation subtypes are used for Pāli compounds to show how the compound's parts relate to each other. + +A Pāli compound is identified through the fact that only the final member of the compound is grammatically inflected, and the earlier part(s) of the compound are in their uninflected form(s). +For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form and only "fallen" *patito* declined. + +Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is in traditional (Pāṇinian) analysis: + + * [compound:dg]() — *dvigu* (a numerical relation) + +~~~ sdparse +satta-ratanāni \n seven jewels + +compound:dg{ratanāni, satta-} + +~~~ + +Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. + +"Compounds" that are only compounds because they are *na-* negated should **not** be treated as compounds. +If the *na-* is in "n'" form (on a verb, as in "*nānubhava*" "doesn't accomplish"), that should be treated as sandhi (as above). +If the *na-* is in "a-" prefix (on a adjective/participle, as in "*amakkhito*" "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. +If, however, the negated word is a compound even without the negation, as in "*anindārosa*" "free of blaming and offense", then the *na* should be treated as a compound part and given the `compound:na` relation in that case. From b5330683b39438ca8d8bad0c2162301f3a76c8a8 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:48:05 +0200 Subject: [PATCH 18/42] Clarify obl:agent usage in documentation Updated the example in the obl-agent documentation to clarify the usage of the obl:agent function. --- _pi/dep/obl-agent.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/_pi/dep/obl-agent.md b/_pi/dep/obl-agent.md index 27ff9457b4c..d7c50d4a821 100644 --- a/_pi/dep/obl-agent.md +++ b/_pi/dep/obl-agent.md @@ -8,9 +8,8 @@ udver: '2' Used for a noun expressing the agent of an action. ~~~ sdparse -Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena -'And the mother of a bodhisattva will be violated by a man of lustful intent.' +Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena \n The mother of bodhisattva cannot be violated by a man of lustful intent -obl:agent (purisena) +obl:agent(anatikkamanīyā, purisena) ~~~ From 6383bc497921b973ecbedabe0a12ebbedb5af69e Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:52:13 +0200 Subject: [PATCH 19/42] Add example for agent expression in obl-agent.md Added example usage for the agent expression in the document. --- _pi/dep/obl-agent.md | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/_pi/dep/obl-agent.md b/_pi/dep/obl-agent.md index d7c50d4a821..6f987f391b8 100644 --- a/_pi/dep/obl-agent.md +++ b/_pi/dep/obl-agent.md @@ -7,9 +7,12 @@ udver: '2' Used for a noun expressing the agent of an action. -~~~ sdparse -Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena \n The mother of bodhisattva cannot be violated by a man of lustful intent +Example: +Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena +'The mother of a bodhisattva cannot be violated by a man of lustful intent.' +~~~ sdparse +Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena obl:agent(anatikkamanīyā, purisena) ~~~ From 5df51368644bf2386150154b8e542f158f8e79b7 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:55:52 +0200 Subject: [PATCH 20/42] Revise 'compound' relation to 'compound:dg' MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Updated the title and short definition for the 'compound' relation. Removed detailed explanations and examples related to Pāli compounds. --- _pi/dep/compound-dg.md | 22 ++++------------------ 1 file changed, 4 insertions(+), 18 deletions(-) diff --git a/_pi/dep/compound-dg.md b/_pi/dep/compound-dg.md index e8881232227..2a7990f21d1 100644 --- a/_pi/dep/compound-dg.md +++ b/_pi/dep/compound-dg.md @@ -1,29 +1,15 @@ --- layout: relation -title: 'compound' -shortdef: 'compound' +title: 'compound:dg' +shortdef: 'dvigu compound' udver: '2' --- - -The `compound` relation subtypes are used for Pāli compounds to show how the compound's parts relate to each other. - -A Pāli compound is identified through the fact that only the final member of the compound is grammatically inflected, and the earlier part(s) of the compound are in their uninflected form(s). -For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form and only "fallen" *patito* declined. - -Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is in traditional (Pāṇinian) analysis: - - * [compound:dg]() — *dvigu* (a numerical relation) +Dvigu compound: compounds whose first member is a numeral ~~~ sdparse -satta-ratanāni \n seven jewels +satta- ratanāni \n seven jewels compound:dg{ratanāni, satta-} ~~~ -Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. - -"Compounds" that are only compounds because they are *na-* negated should **not** be treated as compounds. -If the *na-* is in "n'" form (on a verb, as in "*nānubhava*" "doesn't accomplish"), that should be treated as sandhi (as above). -If the *na-* is in "a-" prefix (on a adjective/participle, as in "*amakkhito*" "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. -If, however, the negated word is a compound even without the negation, as in "*anindārosa*" "free of blaming and offense", then the *na* should be treated as a compound part and given the `compound:na` relation in that case. From 4b7ac3adbb6f556e46e1fa73019c2b9ee0d8f797 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:56:55 +0200 Subject: [PATCH 21/42] Update obl-goal.md --- _pi/dep/obl-goal.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/_pi/dep/obl-goal.md b/_pi/dep/obl-goal.md index b15b70d4bb1..ea3ee58b90d 100644 --- a/_pi/dep/obl-goal.md +++ b/_pi/dep/obl-goal.md @@ -9,8 +9,7 @@ Used for a noun expressing the goal or direction of a movement. We wish to go to your dwellings (RV 1.154.6a). ~~~ sdparse -bodhisatto ... **mātukucchiṃ** okkami -bodhisattva-NOM.M.SG ... mother-womb-AKK.SG.M +bodhisatto mātukucchiṃ okkami \n bodhisattva-NOM.M.SG mother-womb-AKK.SG.M descend.PST.3SG 'Bodisattva descended into the womb of his mother.' obl:goal(mātukucchiṃ) From 99524db20d69919128eac9f0400508ce42f26bf4 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 09:59:11 +0200 Subject: [PATCH 22/42] Create compound-dv.md for dvandva compound Add documentation for dvandva compound in compound-dv.md --- _pi/dep/compound-dv.md | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 _pi/dep/compound-dv.md diff --git a/_pi/dep/compound-dv.md b/_pi/dep/compound-dv.md new file mode 100644 index 00000000000..a8ef88cb2ef --- /dev/null +++ b/_pi/dep/compound-dv.md @@ -0,0 +1,13 @@ +--- +layout: relation +title: 'compound:dv' +shortdef: 'dvandva compound' +udver: '2' +--- +Dvandva compound: compounds that express a coordinative relation + +~~~ sdparse +candima- sūriyā \n moon sun +compound:dg{sūriyā, candima-} +~~~ + From ead6a771ac7cd81b27321c1aa880a2193bc9d195 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:01:47 +0200 Subject: [PATCH 23/42] Change title and definition for compound:tp Updated title and short definition for compound type. --- _pi/dep/compound-tp | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 _pi/dep/compound-tp diff --git a/_pi/dep/compound-tp b/_pi/dep/compound-tp new file mode 100644 index 00000000000..e468f3a6fd2 --- /dev/null +++ b/_pi/dep/compound-tp @@ -0,0 +1,13 @@ +--- +layout: relation +title: 'compound:tp' +shortdef: 'tatpuruṣa compound' +udver: '2' +--- +Tatpuruṣa compound: ompounds whose first member is in an oblique relationship with the second member + +~~~ sdparse +bodhisatta- mātā \n bodhisattva mother +compound:tp{mātā, bodhisatta-} +~~~ + From d62cb339ac8b28ec8e3a0c869411cd0d783166a3 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:02:00 +0200 Subject: [PATCH 24/42] Rename compound-tp to compound-tp.md --- _pi/dep/{compound-tp => compound-tp.md} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename _pi/dep/{compound-tp => compound-tp.md} (100%) diff --git a/_pi/dep/compound-tp b/_pi/dep/compound-tp.md similarity index 100% rename from _pi/dep/compound-tp rename to _pi/dep/compound-tp.md From 7e273565e25d9b241b175ba05f40791e4be35a89 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 15:01:22 +0700 Subject: [PATCH 25/42] add parse for example --- _pi/dep/obl-agent.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/_pi/dep/obl-agent.md b/_pi/dep/obl-agent.md index 6f987f391b8..c1bced2ab73 100644 --- a/_pi/dep/obl-agent.md +++ b/_pi/dep/obl-agent.md @@ -12,7 +12,8 @@ Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena 'The mother of a bodhisattva cannot be violated by a man of lustful intent.' ~~~ sdparse -Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena -obl:agent(anatikkamanīyā, purisena) +Anatikkamanīyā bodhisattamātā hoti kenaci purisena rattacittena \n Not-violatable Bodhisattva-mother is by-any by-man with-lustful-intent +obl:agent(Anatikkamanīyā, purisena) +obl:agent(Not-violatable, by-man) ~~~ From 9199f8828f31e2f24201ca050b94bc86f5229a6d Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 15:03:25 +0700 Subject: [PATCH 26/42] remove redundancy --- _pi/dep/obl-agent.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/_pi/dep/obl-agent.md b/_pi/dep/obl-agent.md index c1bced2ab73..00193c84c6c 100644 --- a/_pi/dep/obl-agent.md +++ b/_pi/dep/obl-agent.md @@ -7,10 +7,6 @@ udver: '2' Used for a noun expressing the agent of an action. -Example: -Anatikkamanīyā ca bodhisattamātā hoti kenaci purisena rattacittena -'The mother of a bodhisattva cannot be violated by a man of lustful intent.' - ~~~ sdparse Anatikkamanīyā bodhisattamātā hoti kenaci purisena rattacittena \n Not-violatable Bodhisattva-mother is by-any by-man with-lustful-intent obl:agent(Anatikkamanīyā, purisena) From 273ecedd14c0ec3fd827fa3a183b671cabb18dd3 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:10:07 +0200 Subject: [PATCH 27/42] Create compound-kd.md --- _pi/dep/compound-kd.md | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 _pi/dep/compound-kd.md diff --git a/_pi/dep/compound-kd.md b/_pi/dep/compound-kd.md new file mode 100644 index 00000000000..ad4b62fbd16 --- /dev/null +++ b/_pi/dep/compound-kd.md @@ -0,0 +1,13 @@ +--- +layout: relation +title: 'compound:kd' +shortdef: 'karmadhāraya compound' +udver: '2' +--- +karmadhāraya compound: compounds whose first member qualifies the second member + +~~~ sdparse +khīṇā- āsavānaṃ \n destroyed inflow-GEN.PL +compound:kd{asavānaṃ, khīṇā-} +~~~ + From 5219ca9b0537e4a9dcdbeaa2877a87d049687a9c Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:10:23 +0200 Subject: [PATCH 28/42] Fix compound notation from 'dg' to 'dv' --- _pi/dep/compound-dv.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/_pi/dep/compound-dv.md b/_pi/dep/compound-dv.md index a8ef88cb2ef..2ab93f1c016 100644 --- a/_pi/dep/compound-dv.md +++ b/_pi/dep/compound-dv.md @@ -8,6 +8,6 @@ Dvandva compound: compounds that express a coordinative relation ~~~ sdparse candima- sūriyā \n moon sun -compound:dg{sūriyā, candima-} +compound:dv{sūriyā, candima-} ~~~ From f7a5391d556ad64bade757ec3d1016ebd661c323 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 15:24:51 +0700 Subject: [PATCH 29/42] more details for dg --- _pi/dep/compound-dg.md | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) diff --git a/_pi/dep/compound-dg.md b/_pi/dep/compound-dg.md index 2a7990f21d1..e0a6dfa80dc 100644 --- a/_pi/dep/compound-dg.md +++ b/_pi/dep/compound-dg.md @@ -4,12 +4,27 @@ title: 'compound:dg' shortdef: 'dvigu compound' udver: '2' --- -Dvigu compound: compounds whose first member is a numeral + +Dvigu compounds are compounds whose first member is a numeral. ~~~ sdparse -satta- ratanāni \n seven jewels +satta- ratanāni \n seven-fold jewels compound:dg{ratanāni, satta-} - +compound:dg({jewels, seven-fold}) ~~~ +The numeral is the dependent and the part of the compound being modified (usually the immediate next token) is the head. + +Note that "hundreds" and "thousands" in Pāli are done in the reverse order, with the number as the substantive and quantified noun in a case relation. For such cases, use the [compound:tp](). + +For cases with both kinds of numbers, use both: + +~~~ sdparse +asīti- vassa- sahassāni \n 80 rains thousands + +compound:dg{sahassāni, asīti-} +compound:tp{sahassāni, vassa-} +compound:dg{thousands, 80} +compound:tp{thousands, rains} +~~~ From 984fac02786031b13c7339f47c364437bdcb8ecd Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:36:11 +0200 Subject: [PATCH 30/42] Create compound-av.md --- _pi/dep/compound-av.md | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 _pi/dep/compound-av.md diff --git a/_pi/dep/compound-av.md b/_pi/dep/compound-av.md new file mode 100644 index 00000000000..d4e29e00105 --- /dev/null +++ b/_pi/dep/compound-av.md @@ -0,0 +1,13 @@ +--- +layout: relation +title: 'compound:av' +shortdef: 'avyayībhāva compound' +udver: '2' +--- +avyayībhāva compound: adverbial compounds whose first member is an indeclinable elements + +~~~ sdparse +evaṃ- maha- ānubhāvā \n thus great power +compound:av{maha-, evaṃ-} +~~~ + From e77a5d89806ab82cba73692e4a1d27b406975fc6 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:38:46 +0200 Subject: [PATCH 31/42] Remove example from compound-av.md --- _pi/dep/compound-av.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/_pi/dep/compound-av.md b/_pi/dep/compound-av.md index d4e29e00105..8e572c979ba 100644 --- a/_pi/dep/compound-av.md +++ b/_pi/dep/compound-av.md @@ -7,7 +7,6 @@ udver: '2' avyayībhāva compound: adverbial compounds whose first member is an indeclinable elements ~~~ sdparse -evaṃ- maha- ānubhāvā \n thus great power -compound:av{maha-, evaṃ-} + ~~~ From f3d9488ebbec04714fdc6a3f6c02180eed7f35db Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 15:42:17 +0700 Subject: [PATCH 32/42] more compound:dv info --- _pi/dep/compound-dv.md | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/_pi/dep/compound-dv.md b/_pi/dep/compound-dv.md index 2ab93f1c016..14c7f618096 100644 --- a/_pi/dep/compound-dv.md +++ b/_pi/dep/compound-dv.md @@ -4,10 +4,25 @@ title: 'compound:dv' shortdef: 'dvandva compound' udver: '2' --- -Dvandva compound: compounds that express a coordinative relation + +Dvandva compounds are those that express a coordinative relation. + +For example "the sun and moon": ~~~ sdparse candima- sūriyā \n moon sun + compound:dv{sūriyā, candima-} +compound:dv{sun, moon} ~~~ +Note that the `compound:dv` relation takes the first element as the dependent and the last element of the coordination as the head. So, in the case of more than two, co-equal members: + +~~~ sdparse +jāti- jarā- maraṇa \n birth aging and-death + +compound:dv{maraṇa, jarā-} +compound:dv{maraṇa, jāti-} +compound:dv{and-death, aging} +compound:dv{and-death, birth} +~~~ From 99c1f66d86a32397cfa92b31f2b7f5b7ea5aec69 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:53:31 +0200 Subject: [PATCH 33/42] =?UTF-8?q?Update=20compound-av.md=20with=20avyay?= =?UTF-8?q?=C4=ABbh=C4=81va=20example?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Added example of avyayībhāva compound with sdparse. --- _pi/dep/compound-av.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/_pi/dep/compound-av.md b/_pi/dep/compound-av.md index 8e572c979ba..b5e092a4a99 100644 --- a/_pi/dep/compound-av.md +++ b/_pi/dep/compound-av.md @@ -7,6 +7,8 @@ udver: '2' avyayībhāva compound: adverbial compounds whose first member is an indeclinable elements ~~~ sdparse +yathā- kamma /n according_to action +compound:av(kamma, yathā-) ~~~ From 15a85cd110dc9104d8b0844e7a9f34a789e6ca6a Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:55:54 +0200 Subject: [PATCH 34/42] Create compound-bv.md --- _pi/dep/compound-bv.md | 13 +++++++++++++ 1 file changed, 13 insertions(+) create mode 100644 _pi/dep/compound-bv.md diff --git a/_pi/dep/compound-bv.md b/_pi/dep/compound-bv.md new file mode 100644 index 00000000000..d55e9d91f31 --- /dev/null +++ b/_pi/dep/compound-bv.md @@ -0,0 +1,13 @@ +--- +layout: relation +title: 'compound:bv' +shortdef: 'bahuvrīhi compound' +udver: '2' +--- +Bahuvrīhi compound: exocentric compounds indicating possessive relationship + +~~~ sdparse +sa- devake /n with god-PL.M.NOM. +compound:bv{devake, sa-} +~~~ + From 0859be129c40db5bd6f072b37b627796f047ef88 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 10:57:14 +0200 Subject: [PATCH 35/42] Fix syntax for adverbial compound notation --- _pi/dep/compound-av.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/_pi/dep/compound-av.md b/_pi/dep/compound-av.md index b5e092a4a99..93a8ca5b976 100644 --- a/_pi/dep/compound-av.md +++ b/_pi/dep/compound-av.md @@ -8,7 +8,7 @@ avyayībhāva compound: adverbial compounds whose first member is an indeclinabl ~~~ sdparse yathā- kamma /n according_to action -compound:av(kamma, yathā-) +compound:av{kamma, yathā-} ~~~ From df79c8dc69947e7c437bbaf4adced08d97aadde6 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 11:08:26 +0200 Subject: [PATCH 36/42] Create compound-na.md --- _pi/dep/compound-na.md | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) create mode 100644 _pi/dep/compound-na.md diff --git a/_pi/dep/compound-na.md b/_pi/dep/compound-na.md new file mode 100644 index 00000000000..7fd908f8dc4 --- /dev/null +++ b/_pi/dep/compound-na.md @@ -0,0 +1,20 @@ +--- +layout: relation +title: 'compound:na' +shortdef: 'nañtatpuruṣa compound' +udver: '2' +--- + +nañtatpuruṣa compounds are those that whose first member is a negation. "Compounds" that are only compounds because they are *na-* negated should **not** be treated as compounds. +If the *na-* is in "n'" form (on a verb, as in "*nānubhava*" "doesn't accomplish"), that should be treated as sandhi (as above). +If the *na-* is in "a-" prefix (on a adjective/participle, as in "*amakkhito*" "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. +If, however, the negated word is a compound even without the negation, as in "*anindārosa*" "free of blaming and offense", then the *na* should be treated as a compound part and given the `compound:na` relation in that case. + +~~~ sdparse +a- nindā- rosa \n NEG blaming offense + +compound:na{nindā-, a-} +compound:dv{nindā-, rosa} +~~~ + + From 599515d6d8b8cf52b18a434c7268208ba1844e03 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 16:09:14 +0700 Subject: [PATCH 37/42] more tp info --- _pi/dep/compound-tp.md | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/_pi/dep/compound-tp.md b/_pi/dep/compound-tp.md index e468f3a6fd2..a409e1c8eb4 100644 --- a/_pi/dep/compound-tp.md +++ b/_pi/dep/compound-tp.md @@ -4,10 +4,24 @@ title: 'compound:tp' shortdef: 'tatpuruṣa compound' udver: '2' --- -Tatpuruṣa compound: ompounds whose first member is in an oblique relationship with the second member + +Tatpuruṣa compounds are compounds whose first member is in an oblique case relationship with its head. For example, a genitive relation: ~~~ sdparse -bodhisatta- mātā \n bodhisattva mother -compound:tp{mātā, bodhisatta-} +bodhisatta- mātā \n the-Bodhisattva's mother + +compound:tp(mātā, bodhisatta-) +compound:tp(mother, the-Bodhisattva's) +~~~ + +Always point the dependent to the head of the relation, which may or may not be the head of the whole compound: + +~~~ sdparse +kāma- guṇ- ūpasaṃhitaṃ \n sensual quality concern + +compound:tp(guṇ-, kāma-) +compound:tp(ūpasaṃhitaṃ, guṇ-) +compound:tp(quality, sensual) +compound:tp(concern, quality) ~~~ From cefbe21b0b57b65bbdc4cc728d37f91ac47f5fcd Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 11:10:47 +0200 Subject: [PATCH 38/42] Update compound annotation guidelines in documentation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Clarified guidelines on treating compounds and negations in Pāli UD. --- _pi/dep/compound.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md index ccc7aa44c49..0fa86bb496d 100644 --- a/_pi/dep/compound.md +++ b/_pi/dep/compound.md @@ -22,7 +22,3 @@ Because the last part of the compound is inflected, we treat that part as the ro Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. -"Compounds" that are only compounds because they are *na-* negated should **not** be treated as compounds. -If the *na-* is in "n'" form (on a verb, as in "*nānubhava*" "doesn't accomplish"), that should be treated as sandhi (as above). -If the *na-* is in "a-" prefix (on a adjective/participle, as in "*amakkhito*" "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. -If, however, the negated word is a compound even without the negation, as in "*anindārosa*" "free of blaming and offense", then the *na* should be treated as a compound part and given the `compound:na` relation in that case. From 5dcfe0f8b730f16dc86ee3f981e77877d61adee1 Mon Sep 17 00:00:00 2001 From: Chia-Wei Lin Date: Mon, 27 Jul 2026 12:02:39 +0200 Subject: [PATCH 39/42] Update syntax section on participles and their roles Clarify the usage of participles in relation to UPOS and their syntactic roles. --- _pi/template-index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/_pi/template-index.md b/_pi/template-index.md index 3a7ef8ed554..827b39080f3 100644 --- a/_pi/template-index.md +++ b/_pi/template-index.md @@ -119,7 +119,7 @@ Participles have the following `FEATS`: ## Syntax -* +Since all the participles have VERB in UPOS, they cannot be used as amod. They are either `acl` if they modify a noun or `advcl` if they modify a verb or a clause. --- **Instruction**: Give criteria for identifying core arguments (subjects and objects), and describe the range of copula constructions in nonverbal clauses. List all subtype relations used. Include links to language-specific relations definitions if any. From b4015424547acfe48bd19bc82725091a54ba2085 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 17:03:06 +0700 Subject: [PATCH 40/42] more info on compound:na --- _pi/dep/compound-na.md | 32 ++++++++++++++++++++++++++------ 1 file changed, 26 insertions(+), 6 deletions(-) diff --git a/_pi/dep/compound-na.md b/_pi/dep/compound-na.md index 7fd908f8dc4..1346ee651a2 100644 --- a/_pi/dep/compound-na.md +++ b/_pi/dep/compound-na.md @@ -5,16 +5,36 @@ shortdef: 'nañtatpuruṣa compound' udver: '2' --- -nañtatpuruṣa compounds are those that whose first member is a negation. "Compounds" that are only compounds because they are *na-* negated should **not** be treated as compounds. -If the *na-* is in "n'" form (on a verb, as in "*nānubhava*" "doesn't accomplish"), that should be treated as sandhi (as above). -If the *na-* is in "a-" prefix (on a adjective/participle, as in "*amakkhito*" "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. -If, however, the negated word is a compound even without the negation, as in "*anindārosa*" "free of blaming and offense", then the *na* should be treated as a compound part and given the `compound:na` relation in that case. +Nañtatpuruṣa compounds are complex compounds whose first member is a negation. + +## When not to use `compound:na` + +Words that would only be compounds because they are *na-* negated should **not** be treated as compounds! + +If the *na-* is in "n'" form (such as on a verb like "*nānubhava*" = "doesn't accomplish"), that should be treated as simple *sandhi*. + +If the *na-* is an "a-" prefix (on a adjective/participle, such as "*amakkhito*" = "undefiled"), then you should treat the entire negated word as a single token and use the `Polarity=Neg` feature to mark the negation. This also holds for e.g. [tatpuruṣa compounds](./compound-tp) whose first part is negated: + +~~~ sdparse +adhamma-/NOUN[Polarity=Neg] savaṇa/NOUN \n false-teaching listening + +compound:tp(savaṇa, adhamma-) +compound:tp(listening, false-teaching) +~~~ + +Despite being a tatpuruṣa with a negation, *adhammasavaṇa* is _not_ a nañtatpuruṣa. + +## When to use `compound:na` + +If, however, the negated word is a compound even without the negation and the negation is applied to the compound as a whole, then the *na* should be treated as a compound part and given the `compound:na` relation: ~~~ sdparse a- nindā- rosa \n NEG blaming offense -compound:na{nindā-, a-} -compound:dv{nindā-, rosa} +compound:na(rosa, a-) +compound:dv(rosa, nindā-) +compound:na(offense, NEG) +compound:dv(offense, blaming) ~~~ From cc1c9f893ef5f16628397146a6f7ec03f2f20c72 Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Mon, 27 Jul 2026 17:26:07 +0700 Subject: [PATCH 41/42] some additional fixes --- _pi/dep/compound-av.md | 9 +++++---- _pi/dep/compound-bv.md | 9 ++++++--- _pi/dep/compound-kd.md | 9 ++++++--- _pi/dep/compound-na.md | 2 +- _pi/dep/compound.md | 4 ++-- _pi/dep/obl-agent.md | 2 +- _pi/dep/obl-goal.md | 1 - 7 files changed, 21 insertions(+), 15 deletions(-) diff --git a/_pi/dep/compound-av.md b/_pi/dep/compound-av.md index 93a8ca5b976..0327e5852f9 100644 --- a/_pi/dep/compound-av.md +++ b/_pi/dep/compound-av.md @@ -4,11 +4,12 @@ title: 'compound:av' shortdef: 'avyayībhāva compound' udver: '2' --- -avyayībhāva compound: adverbial compounds whose first member is an indeclinable elements + +Avyayībhāva compounds are adverbial compounds whose first member is an indeclinable element. ~~~ sdparse -yathā- kamma /n according_to action -compound:av{kamma, yathā-} +yathā- kamma /n according-to action +compound:av(kamma, yathā-) +compound:av(action, according-to) ~~~ - diff --git a/_pi/dep/compound-bv.md b/_pi/dep/compound-bv.md index d55e9d91f31..d8c84c97ea3 100644 --- a/_pi/dep/compound-bv.md +++ b/_pi/dep/compound-bv.md @@ -4,10 +4,13 @@ title: 'compound:bv' shortdef: 'bahuvrīhi compound' udver: '2' --- -Bahuvrīhi compound: exocentric compounds indicating possessive relationship + +Bahuvrīhi compounds are exocentric compounds indicating a possessive relationship: ~~~ sdparse -sa- devake /n with god-PL.M.NOM. -compound:bv{devake, sa-} +sa- devake /n with gods + +compound:bv(devake, sa-) +compound:bv(gods, with) ~~~ diff --git a/_pi/dep/compound-kd.md b/_pi/dep/compound-kd.md index ad4b62fbd16..f94ff48f7b0 100644 --- a/_pi/dep/compound-kd.md +++ b/_pi/dep/compound-kd.md @@ -4,10 +4,13 @@ title: 'compound:kd' shortdef: 'karmadhāraya compound' udver: '2' --- -karmadhāraya compound: compounds whose first member qualifies the second member + +Karmadhāraya compounds are compounds whose first member qualifies the target member: ~~~ sdparse -khīṇā- āsavānaṃ \n destroyed inflow-GEN.PL -compound:kd{asavānaṃ, khīṇā-} +khīṇā- āsavānaṃ \n destroyed inflows + +compound:kd(asavānaṃ, khīṇā-) +compound:kd(inflows, destroyed) ~~~ diff --git a/_pi/dep/compound-na.md b/_pi/dep/compound-na.md index 1346ee651a2..babef53f309 100644 --- a/_pi/dep/compound-na.md +++ b/_pi/dep/compound-na.md @@ -22,7 +22,7 @@ compound:tp(savaṇa, adhamma-) compound:tp(listening, false-teaching) ~~~ -Despite being a tatpuruṣa with a negation, *adhammasavaṇa* is _not_ a nañtatpuruṣa. +Despite being a tatpuruṣa with a negation, *adhammasavaṇa* is not a nañtatpuruṣa. ## When to use `compound:na` diff --git a/_pi/dep/compound.md b/_pi/dep/compound.md index 0fa86bb496d..454987eb06f 100644 --- a/_pi/dep/compound.md +++ b/_pi/dep/compound.md @@ -10,7 +10,7 @@ The `compound` relation subtypes are used for Pāli compounds to show how the co A Pāli compound is identified through the fact that only the final member of the compound is grammatically inflected, and the earlier part(s) of the compound are in their uninflected form(s). For example, *rukkhapatito* "tree-fallen" has *rukkha* "tree" in its stem form and only "fallen" *patito* declined. -Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is in traditional (Pāṇinian) analysis: +Because the last part of the compound is inflected, we treat that part as the root of the compound which relates to the rest of the sentence and the earlier part(s) of the compound receive dependency relations (directly or indirectly) to that final part. The dependency relation to be used for this purpose is `compound` with the specific subtype for the kind of compound it is according to traditional (Pāṇinian) Sanskritic analysis: * [compound:dg]() — *dvigu* (a numerical relation) * [compound:dv]() — *dvandva* (a coordinative relation) @@ -20,5 +20,5 @@ Because the last part of the compound is inflected, we treat that part as the ro * [compound:bv]() - *bahuvrīhi* (a compound used adjectively, e.g. *sa-*) * [compound:av]() — *avyayībhāva* (a compound used adverbially, e.g. *anu-*) -Compounds should always be annotated with a token range (e.g. `10-12`) and token ranges should get the `compound` relations. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. +Compounds should always be annotated with a token range (e.g. `10-12`). The `compound` relations should only be used for such tokens. Sandhi should not be treated as a token range but should be treated instead as two separate tokens with the first token getting `SpaceAfter=No` to mark the lack of a space after it. Pāli UD expresses the difference between sandhi and compounds in this way. diff --git a/_pi/dep/obl-agent.md b/_pi/dep/obl-agent.md index 00193c84c6c..cf292ed619b 100644 --- a/_pi/dep/obl-agent.md +++ b/_pi/dep/obl-agent.md @@ -9,7 +9,7 @@ Used for a noun expressing the agent of an action. ~~~ sdparse Anatikkamanīyā bodhisattamātā hoti kenaci purisena rattacittena \n Not-violatable Bodhisattva-mother is by-any by-man with-lustful-intent + obl:agent(Anatikkamanīyā, purisena) obl:agent(Not-violatable, by-man) ~~~ - diff --git a/_pi/dep/obl-goal.md b/_pi/dep/obl-goal.md index ea3ee58b90d..45841353400 100644 --- a/_pi/dep/obl-goal.md +++ b/_pi/dep/obl-goal.md @@ -15,4 +15,3 @@ bodhisatto mātukucchiṃ okkami \n bodhisattva-NOM.M.SG mother-womb-AKK.SG.M de obl:goal(mātukucchiṃ) ~~~ - From ee6d6d486316ca05d8852ea9e86ddd4101c5adfb Mon Sep 17 00:00:00 2001 From: Khemarato Bhikkhu Date: Tue, 28 Jul 2026 15:13:01 +0700 Subject: [PATCH 42/42] A real index --- _pi/{template-index.md => index.md} | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) rename _pi/{template-index.md => index.md} (93%) diff --git a/_pi/template-index.md b/_pi/index.md similarity index 93% rename from _pi/template-index.md rename to _pi/index.md index 827b39080f3..784c0b2cf42 100644 --- a/_pi/template-index.md +++ b/_pi/index.md @@ -128,14 +128,6 @@ Since all the participles have VERB in UPOS, they cannot be used as amod. They a ## Treebanks -There are [N](../treebanks/pi-comparison.html) Pali UD treebanks: +There is one Pāli UD treebank: - * [Pali-A](../treebanks/pi_a/index.html) - * [Pali-B](../treebanks/pi_b/index.html) - ---- -**Instruction**: Treebank-specific pages are generated automatically from the README file in the treebank repository and -from the data in the latest release. Link to the respective `*-index.html` page in the `treebanks` folder, using the language code -and the treebank code in the file name. - ---- + * [Pali-PaliCanon](https://github.com/UniversalDependencies/UD_Pali-PaliCanon)