Rationalize code placement between wchar.c, encnames.c, and mbutils.c.

Move all the backend-only code that'd crept into wchar.c and encnames.c
into mbutils.c.

To remove the last few #ifdef dependencies from wchar.c and encnames.c,
also make the following changes:

* Adjust get_encoding_name_for_icu to return NULL, not throw an error,
for unsupported encodings.  Its sole caller can perfectly well throw an
error instead.  (While at it, I also made this function and its sibling
is_encoding_supported_by_icu proof against out-of-range encoding IDs.)

* Remove the overlength-name error condition from pg_char_to_encoding.
It's completely silly not to treat that just like any other
the-name-is-not-in-the-table case.

Also, get rid of pg_mic_mblen --- there's no obvious reason why
conv.c shouldn't call pg_mule_mblen instead.

Other than that, this is just code movement and comment-polishing with
no functional changes.  Notably, I reordered declarations in pg_wchar.h
to show which functions are frontend-accessible and which are not.

Discussion: https://postgr.es/m/CA+TgmoYO8oq-iy8E02rD8eX25T-9SmyxKWqqks5OMHxKvGXpXQ@mail.gmail.com

src/backend/utils/adt/pg_locale.c

@@ -1555,9 +1555,14 @@ init_icu_converter(void)
 	UConverter *conv;
 
 	if (icu_converter)
-		return;
+		return;					/* already done */
 
 	icu_encoding_name = get_encoding_name_for_icu(GetDatabaseEncoding());
+	if (!icu_encoding_name)
+		ereport(ERROR,
+				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
+				 errmsg("encoding \"%s\" not supported by ICU",
+						pg_encoding_to_char(GetDatabaseEncoding()))));
 
 	status = U_ZERO_ERROR;
 	conv = ucnv_open(icu_encoding_name, &status);

src/backend/utils/mb/conv.c

@@ -115,7 +115,7 @@ mic2latin(const unsigned char *mic, unsigned char *p, int len,
 		}
 		else
 		{
-			int			l = pg_mic_mblen(mic);
+			int			l = pg_mule_mblen(mic);
 
 			if (len < l)
 				report_invalid_encoding(PG_MULE_INTERNAL, (const char *) mic,
@@ -217,7 +217,7 @@ mic2latin_with_table(const unsigned char *mic,
 		}
 		else
 		{
-			int			l = pg_mic_mblen(mic);
+			int			l = pg_mule_mblen(mic);
 
 			if (len < l)
 				report_invalid_encoding(PG_MULE_INTERNAL, (const char *) mic,

src/common/encnames.c

@@ -10,12 +10,7 @@
  *
  *-------------------------------------------------------------------------
  */
-#ifdef FRONTEND
-#include "postgres_fe.h"
-#else
-#include "postgres.h"
-#include "utils/builtins.h"
-#endif
+#include "c.h"
 
 #include <ctype.h>
 #include <unistd.h>
@@ -310,6 +305,7 @@ static const pg_encname pg_encname_tbl[] =
 #else
 #define DEF_ENC2NAME(name, codepage) { #name, PG_##name, codepage }
 #endif
+
 const pg_enc2name pg_enc2name_tbl[] =
 {
 	DEF_ENC2NAME(SQL_ASCII, 0),
@@ -409,10 +405,8 @@ const pg_enc2gettext pg_enc2gettext_tbl[] =
 };
 
 
-#ifndef FRONTEND
-
 /*
- * Table of encoding names for ICU
+ * Table of encoding names for ICU (currently covers backend encodings only)
  *
  * Reference: <https://ssl.icu-project.org/icu-bin/convexp>
  *
@@ -457,33 +451,32 @@ static const char *const pg_enc2icu_tbl[] =
 	"KOI8-U",					/* PG_KOI8U */
 };
 
+
+/*
+ * Is this encoding supported by ICU?
+ */
 bool
 is_encoding_supported_by_icu(int encoding)
 {
+	if (!PG_VALID_BE_ENCODING(encoding))
+		return false;
 	return (pg_enc2icu_tbl[encoding] != NULL);
 }
 
+/*
+ * Returns ICU's name for encoding, or NULL if not supported
+ */
 const char *
 get_encoding_name_for_icu(int encoding)
 {
-	const char *icu_encoding_name;
-
 	StaticAssertStmt(lengthof(pg_enc2icu_tbl) == PG_ENCODING_BE_LAST + 1,
 					 "pg_enc2icu_tbl incomplete");
 
-	icu_encoding_name = pg_enc2icu_tbl[encoding];
-
-	if (!icu_encoding_name)
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("encoding \"%s\" not supported by ICU",
-						pg_encoding_to_char(encoding))));
-
-	return icu_encoding_name;
+	if (!PG_VALID_BE_ENCODING(encoding))
+		return NULL;
+	return pg_enc2icu_tbl[encoding];
 }
 
-#endif							/* not FRONTEND */
-
 
 /* ----------
  * Encoding checks, for error returns -1 else encoding id
@@ -523,9 +516,10 @@ pg_valid_server_encoding_id(int encoding)
 	return PG_VALID_BE_ENCODING(encoding);
 }
 
-/* ----------
- * Remove irrelevant chars from encoding name
- * ----------
+/*
+ * Remove irrelevant chars from encoding name, store at *newkey
+ *
+ * (Caller's responsibility to provide a large enough buffer)
  */
 static char *
 clean_encoding_name(const char *key, char *newkey)
@@ -547,11 +541,10 @@ clean_encoding_name(const char *key, char *newkey)
 	return newkey;
 }
 
-/* ----------
+/*
  * Search encoding by encoding name
  *
- * Returns encoding ID, or -1 for error
- * ----------
+ * Returns encoding ID, or -1 if not recognized
  */
 int
 pg_char_to_encoding(const char *name)
@@ -568,16 +561,8 @@ pg_char_to_encoding(const char *name)
 		return -1;
 
 	if (strlen(name) >= NAMEDATALEN)
-	{
-#ifdef FRONTEND
-		fprintf(stderr, "encoding name too long\n");
-		return -1;
-#else
-		ereport(ERROR,
-				(errcode(ERRCODE_NAME_TOO_LONG),
-				 errmsg("encoding name too long")));
-#endif
-	}
+		return -1;				/* it's certainly not in the table */
+
 	key = clean_encoding_name(name, buff);
 
 	while (last >= base)
@@ -599,16 +584,6 @@ pg_char_to_encoding(const char *name)
 	return -1;
 }
 
-#ifndef FRONTEND
-Datum
-PG_char_to_encoding(PG_FUNCTION_ARGS)
-{
-	Name		s = PG_GETARG_NAME(0);
-
-	PG_RETURN_INT32(pg_char_to_encoding(NameStr(*s)));
-}
-#endif
-
 const char *
 pg_encoding_to_char(int encoding)
 {
@@ -621,15 +596,3 @@ pg_encoding_to_char(int encoding)
 	}
 	return "";
 }
-
-#ifndef FRONTEND
-Datum
-PG_encoding_to_char(PG_FUNCTION_ARGS)
-{
-	int32		encoding = PG_GETARG_INT32(0);
-	const char *encoding_name = pg_encoding_to_char(encoding);
-
-	return DirectFunctionCall1(namein, CStringGetDatum(encoding_name));
-}
-
-#endif

src/include/mb/pg_wchar.h

@@ -9,7 +9,7 @@
  * src/include/mb/pg_wchar.h
  *
  *	NOTES
- *		This is used both by the backend and by libpq, but should not be
+ *		This is used both by the backend and by frontends, but should not be
  *		included by libpq client programs.  In particular, a libpq client
  *		should not assume that the encoding IDs used by the version of libpq
  *		it's linked to match up with the IDs declared here.
@@ -345,12 +345,6 @@ typedef struct pg_enc2gettext
 
 extern const pg_enc2gettext pg_enc2gettext_tbl[];
 
-/*
- * Encoding names for ICU
- */
-extern bool is_encoding_supported_by_icu(int encoding);
-extern const char *get_encoding_name_for_icu(int encoding);
-
 /*
  * pg_wchar stuff
  */
@@ -539,8 +533,27 @@ extern const char *pg_encoding_to_char(int encoding);
 extern int	pg_valid_server_encoding_id(int encoding);
 
 /*
- * Remaining functions are not considered part of libpq's API, though many
- * of them do exist inside libpq.
+ * These functions are available to frontend code that links with libpgcommon
+ * (in addition to the ones just above).  The constant tables declared
+ * earlier in this file are also available from libpgcommon.
+ */
+extern int	pg_encoding_mblen(int encoding, const char *mbstr);
+extern int	pg_encoding_dsplen(int encoding, const char *mbstr);
+extern int	pg_encoding_verifymb(int encoding, const char *mbstr, int len);
+extern int	pg_encoding_max_length(int encoding);
+extern int	pg_valid_client_encoding(const char *name);
+extern int	pg_valid_server_encoding(const char *name);
+extern bool is_encoding_supported_by_icu(int encoding);
+extern const char *get_encoding_name_for_icu(int encoding);
+
+extern unsigned char *unicode_to_utf8(pg_wchar c, unsigned char *utf8string);
+extern pg_wchar utf8_to_unicode(const unsigned char *c);
+extern bool pg_utf8_islegal(const unsigned char *source, int length);
+extern int	pg_utf_mblen(const unsigned char *s);
+extern int	pg_mule_mblen(const unsigned char *s);
+
+/*
+ * The remaining functions are backend-only.
  */
 extern int	pg_mb2wchar(const char *from, pg_wchar *to);
 extern int	pg_mb2wchar_with_len(const char *from, pg_wchar *to, int len);
@@ -556,18 +569,12 @@ extern int	pg_char_and_wchar_strncmp(const char *s1, const pg_wchar *s2, size_t
 extern size_t pg_wchar_strlen(const pg_wchar *wstr);
 extern int	pg_mblen(const char *mbstr);
 extern int	pg_dsplen(const char *mbstr);
-extern int	pg_encoding_mblen(int encoding, const char *mbstr);
-extern int	pg_encoding_dsplen(int encoding, const char *mbstr);
-extern int	pg_encoding_verifymb(int encoding, const char *mbstr, int len);
-extern int	pg_mule_mblen(const unsigned char *mbstr);
-extern int	pg_mic_mblen(const unsigned char *mbstr);
 extern int	pg_mbstrlen(const char *mbstr);
 extern int	pg_mbstrlen_with_len(const char *mbstr, int len);
 extern int	pg_mbcliplen(const char *mbstr, int len, int limit);
 extern int	pg_encoding_mbcliplen(int encoding, const char *mbstr,
 								  int len, int limit);
 extern int	pg_mbcharcliplen(const char *mbstr, int len, int limit);
-extern int	pg_encoding_max_length(int encoding);
 extern int	pg_database_encoding_max_length(void);
 extern mbcharacter_incrementer pg_database_encoding_character_incrementer(void);
 
@@ -587,12 +594,6 @@ extern int	GetMessageEncoding(void);
 extern int	pg_bind_textdomain_codeset(const char *domainname);
 #endif
 
-extern int	pg_valid_client_encoding(const char *name);
-extern int	pg_valid_server_encoding(const char *name);
-
-extern unsigned char *unicode_to_utf8(pg_wchar c, unsigned char *utf8string);
-extern pg_wchar utf8_to_unicode(const unsigned char *c);
-extern int	pg_utf_mblen(const unsigned char *);
 extern unsigned char *pg_do_encoding_conversion(unsigned char *src, int len,
 												int src_encoding,
 												int dest_encoding);
@@ -647,8 +648,6 @@ extern void mic2latin_with_table(const unsigned char *mic, unsigned char *p,
 								 int len, int lc, int encoding,
 								 const unsigned char *tab);
 
-extern bool pg_utf8_islegal(const unsigned char *source, int length);
-
 #ifdef WIN32
 extern WCHAR *pgwin32_message_to_UTF16(const char *str, int len, int *utf16len);
 #endif