zsv 1.3.0

data/ext/zsv/vendor/zsv-1.3.0/app/utils/jq.c

@@ -0,0 +1,400 @@
+#include <string.h>
+#include <stdlib.h>
+#include <zsv/utils/string.h>
+#include <zsv/utils/writer.h>
+#include <zsv/utils/jq.h>
+#include <jq.h>
+#include <jv.h>
+
+// all jv_ functions defined here consume their jv value
+// consumes:
+//  jv_string_length_bytes
+//  jv_dumpf
+//  jv_dump_string
+//
+// does not consume:
+//  jv_number_value
+//  jv_string_value
+//  jv_get_kind
+
+static int jv_print_scalar_str(jv value, char inside_string, FILE *f, char as_csv) {
+  f = f ? f : stdout;
+  switch (jv_get_kind(value)) {
+  case JV_KIND_STRING: {
+    size_t len = jv_string_length_bytes(jv_copy(value));
+    const char *s = jv_string_value(value);
+    if (!as_csv)
+      fprintf(f, "%.*s", (int)len, s);
+    else {
+      unsigned char *csv = zsv_csv_quote((unsigned char *)s, len, NULL, 0);
+      if (!csv)
+        fprintf(f, "%.*s", (int)len, s);
+      else {
+        if (inside_string)
+          fprintf(f, "%s%.*s", inside_string > 1 ? ";" : "", (int)(strlen((char *)csv) - 2), csv + 1);
+        else
+          fprintf(f, "%s", csv);
+        free(csv);
+      }
+    }
+    jv_free(value);
+    return 1;
+  } break;
+  default:
+    break;
+  }
+  jv_free(value);
+  return 0;
+}
+
+static int jv_print_scalar(jv value, char inside_string, FILE *f, char as_csv) {
+  f = f ? f : stdout;
+  switch (jv_get_kind(value)) {
+  case JV_KIND_INVALID:
+    fprintf(f, "<invalid json>");
+    jv_free(value);
+    return 1;
+  case JV_KIND_NULL:
+    fprintf(f, "null");
+    jv_free(value);
+    return 1;
+  case JV_KIND_TRUE:
+    fprintf(f, "true");
+    jv_free(value);
+    return 1;
+  case JV_KIND_FALSE:
+    fprintf(f, "false");
+    jv_free(value);
+    return 1;
+  case JV_KIND_NUMBER: {
+    char s[64];
+    int n = snprintf(s, sizeof(s), "%lf", jv_number_value(value));
+    if (n > 0 && (size_t)n < sizeof(s))
+      fprintf(f, "%.*s", (int)zsv_strip_trailing_zeros(s, n), s);
+  }
+    jv_free(value);
+    return 1;
+  default:
+    return jv_print_scalar_str(value, inside_string, f, as_csv);
+  }
+}
+
+static void jv_to_csv_aux(jv value, FILE *f, int inside_string) {
+  f = f ? f : stdout;
+  if (!jv_print_scalar(jv_copy(value), inside_string, f, 1)) {
+    switch (jv_get_kind(value)) {
+    case JV_KIND_ARRAY:
+      // flatten
+      if (!inside_string)
+        fprintf(f, "\"");
+      jv_array_foreach(value, i, item) {
+        if (i)
+          fprintf(f, ";");
+        jv_to_csv_aux(item, f, 1);
+      }
+      if (!inside_string)
+        fprintf(f, "\"");
+      break;
+    case JV_KIND_OBJECT:
+      // flatten
+      if (!inside_string)
+        fprintf(f, "\"");
+      jv_object_foreach(value, key, item) {
+        jv_print_scalar(key, 1, f, 1);
+        fprintf(f, ":");
+        jv_to_csv_aux(item, f, 1);
+        fprintf(f, ";");
+      }
+      if (!inside_string)
+        fprintf(f, "\"");
+      break;
+    default:
+      break;
+    }
+  }
+  jv_free(value);
+}
+
+size_t zsv_jq_fwrite1(void *restrict FILE_ptr, const void *restrict buff, size_t len) {
+  return fwrite(buff, len, 1, FILE_ptr);
+}
+
+void jv_to_json_func(jv value, void *ctx) {
+  struct jv_to_json_ctx *data = ctx;
+  if (data->output_started)
+    data->write1(data->ctx, "\n", 1);
+  data->output_started = 1;
+  if (data->write1 == zsv_jq_fwrite1) {
+    if (data->raw_output && jv_get_kind(value) == JV_KIND_STRING)
+      fwrite(jv_string_value(value), 1, jv_string_length_bytes(jv_copy(value)), data->ctx);
+    else
+      jv_dumpf(value, data->ctx, data->flags);
+  } else {
+    // jv_dump_string is memory-inefficient
+    // would be better to create custom dump function that, instead of writing to string buffer,
+    // could directly invoke func()
+    jv jv_s = jv_dump_string(value, data->flags);
+    const char *p = jv_string_value(jv_s);
+    size_t len = jv_string_length_bytes(jv_copy(jv_s));
+    if (len)
+      data->write1(data->ctx, p, len);
+    jv_free(jv_s);
+  }
+}
+
+void jv_to_csv_multi(jv value, void *jv_to_csv_multi_ctx) {
+  struct jv_to_csv_multi_ctx *ctx = jv_to_csv_multi_ctx;
+  FILE *f = ctx->get_file(ctx->ctx);
+  jv_to_csv(value, f);
+}
+
+void jv_to_csv(jv value, void *file) {
+  FILE *f = file;
+  f = f ? f : stdout;
+  if (jv_print_scalar(jv_copy(value), 0, f, 1))
+    fprintf(f, "\n");
+  else {
+    switch (jv_get_kind(value)) {
+    case JV_KIND_ARRAY:
+      jv_array_foreach(value, i, item) {
+        if (i)
+          fprintf(f, ",");
+        jv_to_csv_aux(item, f, 0);
+      }
+      fprintf(f, "\n");
+      break;
+    case JV_KIND_OBJECT:
+      jv_object_foreach(value, key, item) {
+        jv_print_scalar(key, 0, f, 1);
+        fprintf(f, ",");
+        jv_to_csv_aux(item, f, 0);
+        fprintf(f, "\n");
+      }
+      fprintf(f, "\n");
+      break;
+    default:
+      break;
+    }
+  }
+  jv_free(value);
+}
+
+static void jv_to_txt_aux(jv value, FILE *f) {
+  f = f ? f : stdout;
+  if (!jv_print_scalar(jv_copy(value), 0, f, 0)) {
+    switch (jv_get_kind(value)) {
+    case JV_KIND_ARRAY:
+      // flatten
+      fprintf(f, "[");
+      jv_array_foreach(value, i, item) {
+        if (i)
+          fprintf(f, ";");
+        jv_to_txt_aux(item, f);
+      }
+      fprintf(f, "]");
+      break;
+    case JV_KIND_OBJECT:
+      // flatten
+      fprintf(f, "{");
+      jv_object_foreach(value, key, item) {
+        jv_print_scalar(key, 1, f, 0);
+        fprintf(f, ":");
+        jv_to_txt_aux(item, f);
+        fprintf(f, ";");
+      }
+      fprintf(f, "}");
+      break;
+    default:
+      break;
+    }
+  }
+  jv_free(value);
+}
+
+void jv_to_txt(jv value, void *file) {
+  FILE *f = file;
+  f = f ? f : stdout;
+  if (jv_print_scalar(jv_copy(value), 0, f, 0))
+    fprintf(f, "\n");
+  else {
+    switch (jv_get_kind(value)) {
+    case JV_KIND_ARRAY:
+      jv_array_foreach(value, i, item) {
+        if (i)
+          fprintf(f, ",");
+        jv_to_txt_aux(item, f);
+      }
+      fprintf(f, "\n");
+      break;
+    case JV_KIND_OBJECT:
+      jv_object_foreach(value, key, item) {
+        jv_print_scalar(key, 0, f, 0);
+        fprintf(f, ",");
+        jv_to_txt_aux(item, f);
+        fprintf(f, "\n");
+      }
+      fprintf(f, "\n");
+      break;
+    default:
+      break;
+    }
+  }
+  jv_free(value);
+}
+
+void jv_to_lqjq(jv value, void *h) {
+  zsv_jq_handle lqjq = h;
+
+  // jv_dump_string is memory-inefficient
+  // would be better to create custom dump function that, instead of writing to string buffer,
+  // could directly invoke func()
+  jv jv_s = jv_dump_string(value, 0);
+  size_t len = jv_string_length_bytes(jv_copy(jv_s));
+  const char *p = jv_string_value(jv_s);
+  if (len)
+    zsv_jq_parse(lqjq, p, len);
+  jv_free(jv_s);
+}
+
+struct zsv_jq_data {
+  void *jq;
+  struct jv_parser *parser;
+  void (*func)(jv, void *);
+  void *ctx;
+
+  FILE *trace;
+  enum zsv_jq_status status;
+  unsigned char non_null : 1;
+  unsigned char _ : 7;
+};
+
+static zsv_jq_handle zsv_jq_new_aux(const unsigned char *filter, void (*func)(jv, void *), void *ctx,
+                                    enum zsv_jq_status *statusp, int init_flags) {
+  enum zsv_jq_status status = zsv_jq_status_ok;
+  struct zsv_jq_data *d = calloc(1, sizeof(*d));
+  if (!d || !(d->jq = jq_init()) || !(d->parser = jv_parser_new(init_flags)))
+    status = zsv_jq_status_memory;
+  else if (!jq_compile(d->jq, (const char *)filter))
+    status = d->status = zsv_jq_status_compile;
+  if (status == zsv_jq_status_ok) {
+    d->func = func;
+    d->ctx = ctx;
+  } else {
+    zsv_jq_delete(d);
+    d = NULL;
+  }
+  if (statusp)
+    *statusp = status;
+  return d;
+}
+
+zsv_jq_handle zsv_jq_new(const unsigned char *filter, void (*func)(jv, void *), void *ctx,
+                         enum zsv_jq_status *statusp) {
+  return zsv_jq_new_aux(filter, func, ctx, statusp, 0);
+}
+
+zsv_jq_handle zsv_jq_new_stream(const unsigned char *filter, void (*func)(jv, void *), void *ctx,
+                                enum zsv_jq_status *statusp) {
+  return zsv_jq_new_aux(filter, func, ctx, statusp, JV_PARSE_STREAMING);
+}
+
+void zsv_jq_delete(zsv_jq_handle h) {
+  if (h) {
+    if (h->parser)
+      jv_parser_free(h->parser);
+    if (h->jq)
+      jq_teardown((jq_state **)&h->jq);
+    free(h);
+  }
+}
+
+static int zsv_jq_process(jq_state *jq, jv value, void (*func)(jv, void *), void *ctx);
+
+size_t zsv_jq_write(const char *s, size_t n, size_t m, zsv_jq_handle h) {
+  zsv_jq_parse(h, s, n * m);
+  return n * m;
+}
+
+enum zsv_jq_status zsv_jq_parse_file(zsv_jq_handle h, FILE *f) {
+  char buff[4096];
+  for (size_t bytes_read = fread(buff, 1, sizeof(buff), f); bytes_read && h->status == zsv_jq_status_ok;
+       bytes_read = fread(buff, 1, sizeof(buff), f)) {
+    zsv_jq_parse(h, buff, bytes_read);
+    if (feof(f))
+      break;
+  }
+  return h->status;
+}
+
+enum zsv_jq_status zsv_jq_parse(zsv_jq_handle restrict h, const void *restrict s, size_t len) {
+  if (h->status != zsv_jq_status_ok)
+    return h->status;
+
+  jv_parser_set_buf(h->parser, (const char *)s, len, 1);
+  if (h->trace)
+    fwrite(s, len, 1, h->trace);
+
+  jv value;
+  while (jv_is_valid(value = jv_parser_next(h->parser))) {
+    if (!h->non_null && jv_get_kind(value) != JV_KIND_NULL)
+      h->non_null = 1;
+    zsv_jq_process(h->jq, value, h->func, h->ctx);
+  }
+
+  jv msg = jv_invalid_get_msg(value);
+  if (jv_get_kind(msg) == JV_KIND_STRING) {
+    fprintf(stderr, "jq: parse error: %s\n", jv_string_value(msg));
+    h->status = zsv_jq_status_error;
+  } else
+    jv_free(msg);
+  jv_free(value);
+
+  return h->status;
+}
+
+void zsv_jq_trace(zsv_jq_handle h, FILE *trace) {
+  if (h)
+    h->trace = trace;
+}
+
+enum zsv_jq_status zsv_jq_finish(zsv_jq_handle h) {
+  jv value;
+  jv_parser_set_buf(h->parser, "", 0, 0);
+  while (jv_is_valid(value = jv_parser_next(h->parser)))
+    zsv_jq_process(h->jq, value, h->func, h->ctx);
+  return zsv_jq_status_ok;
+}
+
+static int zsv_jq_process(jq_state *jq,
+                          jv value, // will be consumed
+                          void (*func)(jv, void *), void *ctx) {
+  int ret = 14;           // No valid results && -e -> exit(4)
+  jq_start(jq, value, 0); // consumes value
+
+  jv result;
+  while (jv_is_valid(result = jq_next(jq))) {
+    ret = 0;
+    func(result, ctx);
+  }
+  jv_free(result);
+
+  return ret;
+}
+
+void jv_to_bool(jv value, void *char_result) {
+  char *c = char_result;
+  switch (jv_get_kind(value)) {
+  case JV_KIND_TRUE:
+    *c = 1;
+    break;
+  case JV_KIND_FALSE:
+    *c = 0;
+    break;
+  case JV_KIND_NUMBER:
+    *c = (jv_number_value(value) > 0);
+    break;
+  default:
+    break;
+  }
+  jv_free(value);
+}

data/ext/zsv/vendor/zsv-1.3.0/app/utils/json.c

@@ -0,0 +1,120 @@
+/*
+ * Copyright (C) 2021 Liquidaty and the zsv/lib contributors
+ * All rights reserved
+ *
+ * This file is part of zsv/lib, distributed under the license defined at
+ * https://opensource.org/licenses/MIT
+ */
+
+#include <stdlib.h>
+#include <string.h>
+#include <stddef.h>
+
+static inline char is_json_esc_char(unsigned char c) {
+  return (c == '"' || c == '\\' || c < 32);
+}
+
+static void str2hex(unsigned char *to, const unsigned char *p, size_t len) {
+  static const char *hex = "0123456789abcdef";
+
+  for (; len--; p++) {
+    *to++ = hex[p[0] >> 4];
+    *to++ = hex[p[0] & 0x0f];
+  }
+}
+
+static inline unsigned json_esc_char(unsigned char c, char replace[]) {
+  unsigned replacelen;
+  char hex = 0;
+  switch (c) {
+  case '"':
+    replace[1] = '"';
+    break;
+  case '\\':
+    replace[1] = '\\';
+    break;
+  case '\b':
+    replace[1] = 'b';
+    break;
+  case '\f':
+    replace[1] = 'f';
+    break;
+  case '\n':
+    replace[1] = 'n';
+    break;
+  case '\r':
+    replace[1] = 'r';
+    break;
+  case '\t':
+    replace[1] = 't';
+    break;
+  default:
+    hex = 1;
+  }
+
+  replace[0] = '\\';
+  if (hex) {
+    // unicode-hex: but 2/3 are not always zeroes...
+    replace[1] = 'u';
+    replace[2] = '0';
+    replace[3] = '0';
+    str2hex((unsigned char *)replace + 4, &c, 1);
+    replacelen = 6;
+    replace[6] = '\0';
+  } else {
+    replacelen = 2;
+    replace[2] = '\0';
+  }
+  return replacelen;
+}
+
+static unsigned json_escaped_str_len(const unsigned char *s, size_t len) {
+  unsigned count = 0;
+  for (size_t i = 0; i < len; i++, count++) {
+    switch (s[i]) {
+    case '"':
+    case '\\':
+    case '\b':
+    case '\f':
+    case '\n':
+    case '\r':
+    case '\t':
+      count++;
+      break;
+    default:
+      if (s[i] < 31)
+        count += 6;
+      break;
+    }
+  }
+  return count + 2; // + 2 for surrounding quotation marks
+}
+
+unsigned char *zsv_json_from_str_n(const unsigned char *s, size_t len) {
+  size_t new_len = json_escaped_str_len(s, len);
+  unsigned char *new_s = calloc(new_len + 2, sizeof(*new_s));
+  if (new_s) {
+    new_s[0] = new_s[new_len - 1] = (unsigned char)'"';
+    if (new_len == len + 2)
+      memcpy(new_s + 1, s, len);
+    else {
+      char replace[8];
+      for (size_t i = 0, j = 1; i < len && j < new_len - 1; i++) {
+        if (!is_json_esc_char(s[i]))
+          new_s[j++] = s[i];
+        else {
+          size_t rlen = json_esc_char(s[i], replace);
+          memcpy(new_s + j, replace, rlen);
+          j += rlen;
+        }
+      }
+    }
+  }
+  return new_s;
+}
+
+unsigned char *zsv_json_from_str(const unsigned char *s) {
+  if (!s)
+    return (unsigned char *)strdup("null");
+  return zsv_json_from_str_n(s, strlen((const char *)s));
+}

data/ext/zsv/vendor/zsv-1.3.0/app/utils/mem.c

@@ -0,0 +1,18 @@
+/*
+ * Copyright (C) 2021 Liquidaty and the zsv/lib contributors
+ * All rights reserved
+ *
+ * This file is part of zsv/lib, distributed under the license defined at
+ * https://opensource.org/licenses/MIT
+ */
+
+#include <stdlib.h> // calloc
+#include <string.h> // memcpy
+
+/* zsv_memdup(): return copy with double-NULL terminator. caller must free() */
+void *zsv_memdup(const void *src, size_t n) {
+  void *m = calloc(1, n + 2);
+  if (n)
+    memcpy(m, src, n);
+  return m;
+}

data/ext/zsv/vendor/zsv-1.3.0/app/utils/memmem.c

@@ -0,0 +1,132 @@
+// from musl 1.37.5 system/lib/libc/musl/src/string/memmem.c
+// clang-format off
+#include <string.h>
+#include <stdint.h>
+static char *twobyte_memmem(const unsigned char *h, size_t k, const unsigned char *n)
+{
+	uint16_t nw = n[0]<<8 | n[1], hw = h[0]<<8 | h[1];
+	for (h++, k--; k; k--, hw = hw<<8 | *++h)
+		if (hw == nw) return (char *)h-1;
+	return 0;
+}
+static char *threebyte_memmem(const unsigned char *h, size_t k, const unsigned char *n)
+{
+	uint32_t nw = n[0]<<24 | n[1]<<16 | n[2]<<8;
+	uint32_t hw = h[0]<<24 | h[1]<<16 | h[2]<<8;
+	for (h+=2, k-=2; k; k--, hw = (hw|*++h)<<8)
+		if (hw == nw) return (char *)h-2;
+	return 0;
+}
+static char *fourbyte_memmem(const unsigned char *h, size_t k, const unsigned char *n)
+{
+	uint32_t nw = n[0]<<24 | n[1]<<16 | n[2]<<8 | n[3];
+	uint32_t hw = h[0]<<24 | h[1]<<16 | h[2]<<8 | h[3];
+	for (h+=3, k-=3; k; k--, hw = hw<<8 | *++h)
+		if (hw == nw) return (char *)h-3;
+	return 0;
+}
+#define MAX(a,b) ((a)>(b)?(a):(b))
+#define MIN(a,b) ((a)<(b)?(a):(b))
+#define BITOP(a,b,op) \
+ ((a)[(size_t)(b)/(8*sizeof *(a))] op (size_t)1<<((size_t)(b)%(8*sizeof *(a))))
+static char *twoway_memmem(const unsigned char *h, const unsigned char *z, const unsigned char *n, size_t l)
+{
+	size_t i, ip, jp, k, p, ms, p0, mem, mem0;
+	size_t byteset[32 / sizeof(size_t)] = { 0 };
+	size_t shift[256];
+	/* Computing length of needle and fill shift table */
+	for (i=0; i<l; i++)
+		BITOP(byteset, n[i], |=), shift[n[i]] = i+1;
+	/* Compute maximal suffix */
+	ip = -1; jp = 0; k = p = 1;
+	while (jp+k<l) {
+		if (n[ip+k] == n[jp+k]) {
+			if (k == p) {
+				jp += p;
+				k = 1;
+			} else k++;
+		} else if (n[ip+k] > n[jp+k]) {
+			jp += k;
+			k = 1;
+			p = jp - ip;
+		} else {
+			ip = jp++;
+			k = p = 1;
+		}
+	}
+	ms = ip;
+	p0 = p;
+	/* And with the opposite comparison */
+	ip = -1; jp = 0; k = p = 1;
+	while (jp+k<l) {
+		if (n[ip+k] == n[jp+k]) {
+			if (k == p) {
+				jp += p;
+				k = 1;
+			} else k++;
+		} else if (n[ip+k] < n[jp+k]) {
+			jp += k;
+			k = 1;
+			p = jp - ip;
+		} else {
+			ip = jp++;
+			k = p = 1;
+		}
+	}
+	if (ip+1 > ms+1) ms = ip;
+	else p = p0;
+	/* Periodic needle? */
+	if (memcmp(n, n+p, ms+1)) {
+		mem0 = 0;
+		p = MAX(ms, l-ms-1) + 1;
+	} else mem0 = l-p;
+	mem = 0;
+	/* Search loop */
+	for (;;) {
+		/* If remainder of haystack is shorter than needle, done */
+		if (z-h < l) return 0;
+		/* Check last byte first; advance by shift on mismatch */
+		if (BITOP(byteset, h[l-1], &)) {
+			k = l-shift[h[l-1]];
+			if (k) {
+				if (mem0 && mem && k < p) k = l-p;
+				h += k;
+				mem = 0;
+				continue;
+			}
+		} else {
+			h += l;
+			mem = 0;
+			continue;
+		}
+		/* Compare right half */
+		for (k=MAX(ms+1,mem); k<l && n[k] == h[k]; k++);
+		if (k < l) {
+			h += k-ms;
+			mem = 0;
+			continue;
+		}
+		/* Compare left half */
+		for (k=ms+1; k>mem && n[k-1] == h[k-1]; k--);
+		if (k <= mem) return (char *)h;
+		h += p;
+		mem = mem0;
+	}
+}
+void *zsvmemmem(const void *h0, size_t k, const void *n0, size_t l)
+{
+	const unsigned char *h = h0, *n = n0;
+	/* Return immediately on empty needle */
+	if (!l) return (void *)h;
+	/* Return immediately when needle is longer than haystack */
+	if (k<l) return 0;
+	/* Use faster algorithms for short needles */
+	h = memchr(h0, *n, k);
+	if (!h || l==1) return (void *)h;
+	k -= h - (const unsigned char *)h0;
+	if (k<l) return 0;
+	if (l==2) return twobyte_memmem(h, k, n);
+	if (l==3) return threebyte_memmem(h, k, n);
+	if (l==4) return fourbyte_memmem(h, k, n);
+	return twoway_memmem(h, h+k, n, l);
+}