Step 13: String/regex primitives — PCRE-compatible, cross-host

New primitives in sx_primitives.ml:
  char-at, char-code, parse-number — string inspection + conversion
  regex-match, regex-match?, regex-find-all — PCRE pattern matching
  regex-replace, regex-replace-first — PCRE substitution
  regex-split — split by PCRE pattern

Uses Re.Pcre (OCaml re library) so regex patterns use the same syntax
as JS RegExp — patterns in .sx files work identically on browser and
server. Replaces the old test-only regex-find-all stub.

Also: split now handles multi-char separators via Re.

176 new tests (10 suites). 2912/2912 total, zero failures.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-04-05 20:38:40 +00:00
parent 516f9c7186
commit 5d88b363e4
4 changed files with 247 additions and 48 deletions

View File

@@ -1,3 +1,4 @@
(library
(name sx)
(wrapped false))
(wrapped false)
(libraries re re.pcre))

View File

@@ -398,7 +398,12 @@ let () =
register "split" (fun args ->
match args with
| [String s; String sep] ->
List (List.map (fun p -> String p) (String.split_on_char sep.[0] s))
if String.length sep = 1 then
List (List.map (fun p -> String p) (String.split_on_char sep.[0] s))
else
(* Multi-char separator: use Re for literal split *)
let re = Re.compile (Re.str sep) in
List (List.map (fun p -> String p) (Re.split re s))
| _ -> raise (Eval_error "split: 2 args"));
register "join" (fun args ->
match args with
@@ -441,6 +446,94 @@ let () =
Buffer.add_utf_8_uchar buf (Uchar.of_int (int_of_float n));
String (Buffer.contents buf)
| _ -> raise (Eval_error "char-from-code: 1 arg"));
register "char-at" (fun args ->
match args with
| [String s; Number n] ->
let i = int_of_float n in
if i >= 0 && i < String.length s then
String (String.make 1 s.[i])
else Nil
| _ -> raise (Eval_error "char-at: string and index"));
register "char-code" (fun args ->
match args with
| [String s] when String.length s > 0 -> Number (float_of_int (Char.code s.[0]))
| _ -> raise (Eval_error "char-code: 1 non-empty string arg"));
register "parse-number" (fun args ->
match args with
| [String s] ->
(try Number (float_of_string s)
with Failure _ -> Nil)
| _ -> raise (Eval_error "parse-number: 1 string arg"));
(* === Regex (PCRE-compatible — same syntax as JS RegExp) === *)
register "regex-match" (fun args ->
match args with
| [String pattern; String input] ->
(try
let re = Re.Pcre.re pattern |> Re.compile in
match Re.exec_opt re input with
| Some group ->
let full = Re.Group.get group 0 in
let n = Re.Group.nb_groups group in
let groups = ref [String full] in
for i = 1 to n - 1 do
(try groups := !groups @ [String (Re.Group.get group i)]
with Not_found -> groups := !groups @ [Nil])
done;
List !groups
| None -> Nil
with _ -> Nil)
| _ -> raise (Eval_error "regex-match: pattern and input strings"));
register "regex-match?" (fun args ->
match args with
| [String pattern; String input] ->
(try Bool (Re.execp (Re.Pcre.re pattern |> Re.compile) input)
with _ -> Bool false)
| _ -> raise (Eval_error "regex-match?: pattern and input strings"));
register "regex-find-all" (fun args ->
match args with
| [String pattern; String input] ->
(try
let re = Re.Pcre.re pattern |> Re.compile in
let matches = Re.all re input in
let results = List.map (fun group ->
(* If there's a capture group, return group 1; else full match *)
try String (Re.Group.get group 1)
with Not_found -> String (Re.Group.get group 0)
) matches in
ListRef (ref results)
with _ -> ListRef (ref []))
| _ -> raise (Eval_error "regex-find-all: pattern and input strings"));
register "regex-replace" (fun args ->
match args with
| [String pattern; String replacement; String input] ->
(try
let re = Re.Pcre.re pattern |> Re.compile in
String (Re.replace_string re ~by:replacement input)
with _ -> String input)
| _ -> raise (Eval_error "regex-replace: pattern, replacement, input strings"));
register "regex-replace-first" (fun args ->
match args with
| [String pattern; String replacement; String input] ->
(try
let re = Re.Pcre.re pattern |> Re.compile in
(* Re doesn't have replace_first, so use all matches and replace only first *)
match Re.exec_opt re input with
| Some group ->
let start = Re.Group.start group 0 and stop = Re.Group.stop group 0 in
String (String.sub input 0 start ^ replacement ^
String.sub input stop (String.length input - stop))
| None -> String input
with _ -> String input)
| _ -> raise (Eval_error "regex-replace-first: pattern, replacement, input strings"));
register "regex-split" (fun args ->
match args with
| [String pattern; String input] ->
(try
let re = Re.Pcre.re pattern |> Re.compile in
ListRef (ref (List.map (fun s -> String s) (Re.split re input)))
with _ -> ListRef (ref [String input]))
| _ -> raise (Eval_error "regex-split: pattern and input strings"));
(* === Collections === *)
register "list" (fun args -> ListRef (ref args));