From 6cc95f664e7227d6b2f9b32580d769299bc28cb7 Mon Sep 17 00:00:00 2001
From: Thomas Hallgren <th-github@altocumulus.org>
Date: Mon, 29 Jan 2018 11:28:11 +0100
Subject: [PATCH 1/3] index.html: update summer school link

---
 index.html | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/index.html b/index.html
index c9129f63c..274998d5a 100644
--- a/index.html
+++ b/index.html
@@ -59,7 +59,7 @@ function sitesearch() {
     <li><A HREF="doc/gf-quickstart.html">QuickStart</A>
     <li><A HREF="doc/gf-reference.html">QuickRefCard</A>
     <li><A HREF="doc/gf-shell-reference.html">GF Shell Reference</A>
-    <li><a href="http://school.grammaticalframework.org/2017/"><b>GF Summer School</b></a>
+    <li><a href="http://school.grammaticalframework.org/"><b>GF Summer School</b></a>
   </ul>
   <ul>
     <li><A HREF="gf-book">The GF Book</A>

From a16fe3415a76a4b101a2cedf96d876fc691b8155 Mon Sep 17 00:00:00 2001
From: Krasimir Angelov <kr.angelov@gmail.com>
Date: Sun, 4 Feb 2018 17:21:15 +0100
Subject: [PATCH 2/3] partial fix for word completion in the C runtime

---
 src/runtime/c/pgf/parser.c | 41 ++++++++++++++++++++++----------------
 1 file changed, 24 insertions(+), 17 deletions(-)

diff --git a/src/runtime/c/pgf/parser.c b/src/runtime/c/pgf/parser.c
index b1916c218..2650e2618 100644
--- a/src/runtime/c/pgf/parser.c
+++ b/src/runtime/c/pgf/parser.c
@@ -65,6 +65,7 @@ typedef enum { BIND_NONE, BIND_HARD, BIND_SOFT } BIND_TYPE;
 typedef struct {
 	PgfProductionIdx* idx;
 	size_t offset;
+	size_t sym_idx;
 } PgfLexiconIdxEntry;
 
 typedef GuBuf PgfLexiconIdx;
@@ -1060,16 +1061,16 @@ pgf_parsing_complete(PgfParsing* ps, PgfItem* item, PgfExprProb *ep)
 }
 
 static int
-pgf_symbols_cmp(GuString* psent, PgfSymbols* syms, bool case_sensitive)
+pgf_symbols_cmp(GuString* psent, PgfSymbols* syms, size_t* sym_idx, bool case_sensitive)
 {
 	size_t n_syms = gu_seq_length(syms);
-	for (size_t i = 0; i < n_syms; i++) {
-		PgfSymbol sym = gu_seq_get(syms, PgfSymbol, i);
+	while (*sym_idx < n_syms) {
+		PgfSymbol sym = gu_seq_get(syms, PgfSymbol, *sym_idx);
 
-		if (i > 0) {
+		if (*sym_idx > 0) {
 			if (!skip_space(psent)) {
 				if (**psent == 0)
-					return -1;
+					return 0;
 				return 1;
 			}
 
@@ -1085,13 +1086,13 @@ pgf_symbols_cmp(GuString* psent, PgfSymbols* syms, bool case_sensitive)
 		case PGF_SYMBOL_LIT:
 		case PGF_SYMBOL_VAR: {
 			if (**psent == 0)
-				return -1;
+				return 0;
 			return 1;
 		}
 		case PGF_SYMBOL_KS: {
 			PgfSymbolKS* pks = inf.data;
 			if (**psent == 0)
-				return -1;
+				return 0;
 
 			int cmp = cmp_string(psent, pks->token, case_sensitive);
 			if (cmp != 0)
@@ -1110,6 +1111,8 @@ pgf_symbols_cmp(GuString* psent, PgfSymbols* syms, bool case_sensitive)
 		default:
 			gu_impossible();
 		}
+
+		(*sym_idx)++;
 	}
 
 	return 0;
@@ -1130,7 +1133,8 @@ pgf_parsing_lookahead(PgfParsing *ps, PgfParseState* state,
 
 		GuString start   = ps->sentence + state->end_offset;
 		GuString current = start;
-		int cmp = pgf_symbols_cmp(&current, seq->syms, ps->case_sensitive);
+		size_t sym_idx = 0;
+		int cmp = pgf_symbols_cmp(&current, seq->syms, &sym_idx, ps->case_sensitive);
 		if (cmp < 0) {
 			j = k-1;
 		} else if (cmp > 0) {
@@ -1151,8 +1155,9 @@ pgf_parsing_lookahead(PgfParsing *ps, PgfParseState* state,
 
 			if (seq->idx != NULL) {
 				PgfLexiconIdxEntry* entry = gu_buf_extend(state->lexicon_idx);
-				entry->idx       = seq->idx;
-				entry->offset    = (size_t) (current - ps->sentence);
+				entry->idx        = seq->idx;
+				entry->offset     = (size_t) (current - ps->sentence);
+				entry->sym_idx    = sym_idx;
 			}
 
 			if (len+1 <= max)
@@ -1278,14 +1283,15 @@ pgf_parsing_add_transition(PgfParsing* ps, PgfToken tok, PgfItem* item)
 static void
 pgf_parsing_predict_lexeme(PgfParsing* ps, PgfItemConts* conts,
                            PgfProductionIdxEntry* entry,
-                           size_t offset)
+                           size_t offset, size_t sym_idx)
 {
 	GuVariantInfo i = { PGF_PRODUCTION_APPLY, entry->papp };
 	PgfProduction prod = gu_variant_close(i);
 	PgfItem* item =
 		pgf_new_item(ps, conts, prod);
 	PgfSymbols* syms = entry->papp->fun->lins[conts->lin_idx]->syms;
-	item->sym_idx = gu_seq_length(syms);
+	item->sym_idx = sym_idx;
+	pgf_item_set_curr_symbol(item, ps->pool);
 	prob_t prob = item->inside_prob+item->conts->outside_prob;
 	PgfParseState* state =
 		pgf_new_parse_state(ps, offset, BIND_NONE, prob);
@@ -1358,7 +1364,7 @@ pgf_parsing_td_predict(PgfParsing* ps,
 									 PgfProductionIdxEntry, &key);
 
 				if (value != NULL) {
-					pgf_parsing_predict_lexeme(ps, conts, value, lentry->offset);
+					pgf_parsing_predict_lexeme(ps, conts, value, lentry->offset, lentry->sym_idx);
 
 					PgfProductionIdxEntry* start =
 						gu_buf_data(lentry->idx);
@@ -1369,7 +1375,7 @@ pgf_parsing_td_predict(PgfParsing* ps,
 					while (left >= start &&
 						   value->ccat->fid == left->ccat->fid &&
 						   value->lin_idx   == left->lin_idx) {
-						pgf_parsing_predict_lexeme(ps, conts, left, lentry->offset);
+						pgf_parsing_predict_lexeme(ps, conts, left, lentry->offset, lentry->sym_idx);
 						left--;
 					}
 
@@ -1377,7 +1383,7 @@ pgf_parsing_td_predict(PgfParsing* ps,
 					while (right <= end &&
 						   value->ccat->fid == right->ccat->fid &&
 						   value->lin_idx   == right->lin_idx) {
-						pgf_parsing_predict_lexeme(ps, conts, right, lentry->offset);
+						pgf_parsing_predict_lexeme(ps, conts, right, lentry->offset, lentry->sym_idx);
 						right++;
 					}
 				}
@@ -2372,8 +2378,9 @@ pgf_sequence_cmp_fn(GuOrder* order, const void* p1, const void* p2)
 	GuString sent = (GuString) p1;
 	const PgfSequence* sp2 = p2;
 
-	int res = pgf_symbols_cmp(&sent, sp2->syms, self->case_sensitive);
-	if (res == 0 && *sent != 0) {
+	size_t sym_idx = 0;
+	int res = pgf_symbols_cmp(&sent, sp2->syms, &sym_idx, self->case_sensitive);
+	if (res == 0 && (*sent != 0 || sym_idx != gu_seq_length(sp2->syms))) {
 		res = 1;
 	}
 

From bb4218433fe8afabbc5239b2ca41731bb3e2da4b Mon Sep 17 00:00:00 2001
From: Krasimir Angelov <kr.angelov@gmail.com>
Date: Thu, 22 Feb 2018 11:35:54 +0100
Subject: [PATCH 3/3] - tweak the tokenizer in pgf_lookup_sentence to threat
 .!?,: as separate tokens + bugfix which causes crashes

---
 src/runtime/c/pgf/lookup.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/src/runtime/c/pgf/lookup.c b/src/runtime/c/pgf/lookup.c
index 21c82450f..5918275c1 100644
--- a/src/runtime/c/pgf/lookup.c
+++ b/src/runtime/c/pgf/lookup.c
@@ -119,7 +119,7 @@ typedef struct {
 static PgfAbsProduction*
 pgf_lookup_new_production(PgfAbsFun* fun, GuPool *pool)
 {
-	size_t n_hypos = gu_seq_length(fun->type->hypos);
+	size_t n_hypos = fun->type->hypos ? gu_seq_length(fun->type->hypos) : 0;
 	PgfAbsProduction* prod = gu_new_flex(pool, PgfAbsProduction, args, n_hypos);
 	prod->fun = fun;
 	prod->count = 0;
@@ -699,8 +699,12 @@ pgf_lookup_tokenize(GuMap* lexicon_idx, GuString sentence, GuPool* pool)
 			break;
 
 		const uint8_t* start = p-1;
-		while (c != 0 && !gu_ucs_is_space(c)) {
+		if (strchr(".!?,:",c) != NULL)
 			c = gu_utf8_decode(&p);
+		else {
+			while (c != 0 && strchr(".!?,:",c) == NULL && !gu_ucs_is_space(c)) {
+				c = gu_utf8_decode(&p);
+			}
 		}
 		const uint8_t* end   = p-1;