initial commit: only applies to select cmd (to do: test, apply to all… (

#50)

* move header offset/depth generic options (-R,--skip-head / -d,--header-row-span) into parser and make available for all commands, instead of only for select command
* other miscellaneous cleanup

app/Makefile

@@ -141,8 +141,8 @@ endif
 
 ZSV=$(BINDIR)/zsv${EXE}
 
-SOURCES= echo count select 2json serialize flatten pretty stack desc 2tsv sql 2db
-CLI_SOURCES=select desc count pretty sql flatten 2json 2tsv serialize stack 2db
+SOURCES= echo count select 2tsv 2json serialize flatten pretty stack desc sql 2db
+CLI_SOURCES=select desc count 2tsv pretty sql flatten 2json serialize stack 2db
 
 ifneq ($(LDFLAGS_JQ),)
   SOURCES+= jq
@@ -210,7 +210,7 @@ help:
 	@echo "which will build and test all apps, or to build/test a single app:"
 	@echo "  ${MAKE} test-xx"
 	@echo "where xx is any of:"
-	@echo "  echo count select 2json serialize flatten pretty stack desc 2tsv sql 2db"
+	@echo "  echo count select 2tsv 2json serialize flatten pretty stack desc sql 2db"
 	@echo ""
 
 install: ${ZSV}

app/benchmark/Makefile

@@ -28,11 +28,11 @@ endif
 CCBN=$(shell basename ${CC})
 ZSVBIN=../../build/${BUILD_SUBDIR}/${CCBN}/bin/zsv_
 
-QUICK=
+QUICK=1
 
 help:
 	@echo "To run all tests (set QUICK to skip mlr and csvcut):"
-	@echo "    make all [QUICK=1]"
+	@echo "    make all [QUICK=0]"
 	@echo "    make CLI"
 
 CLI: ZSVBIN="zsv "

app/builtin/help.c

@@ -43,6 +43,8 @@ static int main_help(int argc, const char *argv[]) {
     "  -t,--tab-delim: set column delimiter to tab",
     "  -O,--other-delim <char>: set column delimiter to specified character",
     "  -q,--no-quote: turn off quote handling",
+    "  -R,--skip-head <n>: skip specified number of initial rows",
+    "  -d,--header-row-span <n>: apply header depth (rowspan) of n",
     "  -v,--verbose: verbose output",
     "",
     "Commands:",

app/ext_example/test/expected/zsvext-test-3.out

@@ -25,6 +25,8 @@ Options common to all commands:
   -t,--tab-delim: set column delimiter to tab
   -O,--other-delim <char>: set column delimiter to specified character
   -q,--no-quote: turn off quote handling
+  -R,--skip-head <n>: skip specified number of initial rows
+  -d,--header-row-span <n>: apply header depth (rowspan) of n
   -v,--verbose: verbose output
 
 Commands:
@@ -68,6 +70,8 @@ Options common to all commands:
   -t,--tab-delim: set column delimiter to tab
   -O,--other-delim <char>: set column delimiter to specified character
   -q,--no-quote: turn off quote handling
+  -R,--skip-head <n>: skip specified number of initial rows
+  -d,--header-row-span <n>: apply header depth (rowspan) of n
   -v,--verbose: verbose output
 
 Commands:

app/jq.c

@@ -2,11 +2,6 @@
 #include <string.h>
 
 #include "jq_internal.h"
-
-size_t fwrite1(void *restrict FILE_ptr, const void *restrict buff, size_t len) {
-  return fwrite(buff, len, 1, FILE_ptr);
-}
-
 #include "jq_internal.c"
 
 #ifndef APPNAME
@@ -82,11 +77,11 @@ int MAIN(int argc, const char *argv[]) {
   if(!err) {
     void (*jqfunc)(jv, void *) = to_csv ? jv_to_csv : jv_to_json_func;
     struct jv_to_json_ctx ctx;
-    ctx.write1 = fwrite1;
+    ctx.write1 = zsv_jq_fwrite1;
     ctx.ctx = f_out;
     ctx.flags = JV_PRINT_PRETTY | JV_PRINT_SPACE1;
 
-    void *jqctx = to_csv ? f_out : &ctx;
+    void *jqctx = to_csv ? (void *)f_out : (void *)&ctx;
     enum zsv_jq_status jqstat;
     zsv_jq_handle zjq = zsv_jq_new(jqfilter, jqfunc, jqctx, &jqstat);
     if(jqstat != zsv_jq_status_ok) {

app/jq_internal.c

@@ -118,21 +118,13 @@ static void jv_to_csv_aux(jv value, FILE *f, int inside_string) {
   jv_free(value);
 }
 
-/*
-void jv_to_json_w_ctx(jv value, void *ctx) {
-  struct jv_to_json_ctx *c = ctx;
-  jv_dumpf(value, ctx->out, c->flags);
+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(jv value, void *file) {
-  FILE *f = file;
-  jv_dumpf(value, f, 0);
-}
-*/
-
 void jv_to_json_func(jv value, void *ctx) {
   struct jv_to_json_ctx *data = ctx;
-  if(data->write1 == fwrite1)
+  if(data->write1 == zsv_jq_fwrite1)
     jv_dumpf(value, data->ctx, data->flags);
   else {
     // jv_dump_string is memory-inefficient
@@ -184,7 +176,6 @@ void jv_to_csv(jv value, void *file) {
   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)) {
@@ -262,8 +253,6 @@ void jv_to_lqjq(jv value, void *h) {
   jv_free(jv_s);
 }
 
-///
-
 struct zsv_jq_data {
   void *jq;
   struct jv_parser *parser;
@@ -398,7 +387,6 @@ static int zsv_jq_process(jq_state *jq,
   return ret;
 }
 
-
 void jv_to_bool(jv value, void *char_result) {
   char *c = char_result;
   switch(jv_get_kind(value)) {
@@ -416,12 +404,3 @@ void jv_to_bool(jv value, void *char_result) {
   }
   jv_free(value);
 }
-
-static const unsigned char *strrchru(const unsigned char *s, char c) {
-  return (const unsigned char *)strrchr((const char *)s, c);
-}
-
-static size_t zsv_jq_parse1(void *restrict h, const void *restrict s, size_t len) {
-  enum zsv_jq_status stat = zsv_jq_parse(h, s, len);
-  return (size_t) stat;
-}

app/jq_internal.h

@@ -11,8 +11,10 @@ enum zsv_jq_status {
   zsv_jq_status_error
 };
 
+size_t zsv_jq_fwrite1(void *restrict FILE_ptr, const void *restrict buff, size_t len);
+
 struct jv_to_json_ctx {
-  size_t (*write1)(void *restrict ctx, const void *restrict buff, size_t len); // e.g. common/write1
+  size_t (*write1)(void *restrict ctx, const void *restrict buff, size_t len); // e.g. zsv_jq_fwrite1
   void *ctx; // e.g. FILE *
   int flags; // passed on to jv_dumpf / jv_dump_string
 };

app/select.c

@@ -83,7 +83,6 @@ struct zsv_select_data {
 
   unsigned int output_col_index; // num of cols printed in current row
   size_t file_row_count;
-  size_t header_rows_processed;
 
   // output columns:
   const char **col_argv;
@@ -112,10 +111,8 @@ struct zsv_select_data {
 
   double sample_pct;
 
-  unsigned char skip_rows;
-  unsigned char skip_rows_orig;
   unsigned char sample_every_n;
-  unsigned char header_depth;
+
   size_t data_rows_limit;
   size_t skip_data_rows;
 
@@ -522,34 +519,27 @@ static void zsv_select_header_row(void *ctx) {
   if(data->cancelled)
     return;
 
-  if(data->skip_rows > 0)
-    data->skip_rows--;
-  else {
-    data->header_rows_processed++;
-    unsigned int cols = zsv_column_count(data->parser);
-    unsigned int max_header_ix = 0;
-    for(unsigned int i = 0; i < cols; i++) {
-      struct zsv_cell cell = zsv_get_cell(data->parser, i);
-      cell.str = zsv_select_cell_clean(data, cell.str, cell.quoted, &cell.len);
-      if(i < data->opts.max_columns) {
-        zsv_select_append_spaced_word(&data->header_names[i], cell.str, cell.len);
-        if(cell.len)
-          max_header_ix = i+1;
-      }
+  unsigned int cols = zsv_column_count(data->parser);
+  unsigned int max_header_ix = 0;
+  for(unsigned int i = 0; i < cols; i++) {
+    struct zsv_cell cell = zsv_get_cell(data->parser, i);
+    cell.str = zsv_select_cell_clean(data, cell.str, cell.quoted, &cell.len);
+    if(i < data->opts.max_columns) {
+      zsv_select_append_spaced_word(&data->header_names[i], cell.str, cell.len);
+      if(cell.len)
+        max_header_ix = i+1;
     }
+  }
 
-    // in case we want to make this an option later
-    char trim_trailing_columns = 1;
-    if(!trim_trailing_columns)
-      max_header_ix = cols;
+  // in case we want to make this an option later
+  char trim_trailing_columns = 1;
+  if(!trim_trailing_columns)
+    max_header_ix = cols;
 
-    if(max_header_ix > data->header_name_count)
-      data->header_name_count = max_header_ix;
+  if(max_header_ix > data->header_name_count)
+    data->header_name_count = max_header_ix;
 
-    // if this was the last row in the header, finish header processing
-    if(data->header_rows_processed >= data->header_depth)
-      zsv_select_header_finish(data);
-  }
+  zsv_select_header_finish(data);
 }
 
 #define ZSV_SELECT_MAX_COLS_DEFAULT 1024
@@ -586,8 +576,6 @@ const char *zsv_select_usage_msg[] =
    "  --distinct: skip subsequent occurrences of columns with the same name",
    "  --merge: merge subsequent occurrences of columns with the same name, outputting first non-null value",
    // --rename: like distinct, but instead of removing cols with dupe names, renames them, trying _<n> for n up to max cols
-   "  -R, --skip-head <n>: skip specified number of rows",
-   "  -D, --skip-data <n>: skip the specified number of data rows",
    "  -e <embedded lineend char>: char to replace embedded lineend. if none provided, embedded lineends are preserved",
    "      If the provided string begins with 0x, it will be interpreted as the hex representation of a string",
    "  -x <column>: exclude the indicated column. can be specified more than once",
@@ -658,8 +646,6 @@ int MAIN(int argc, const char *argv[]) {
     data.opts = zsv_get_default_opts();
 
     struct zsv_csv_writer_options writer_opts = zsv_writer_get_default_opts();
-    data.header_depth = 1;
-
     int col_index_arg_i = 0;
     const char *insert_header_row = NULL;
     for(int arg_i = 1; !err && arg_i < argc; arg_i++) {
@@ -733,13 +719,8 @@ int MAIN(int argc, const char *argv[]) {
       else if(!strcmp(argv[arg_i], "--whitespace-clean-no-newline")) {
         data.clean_white = 1;
         data.whitspace_clean_flags = 1;
-      } else if(!strcmp(argv[arg_i], "-W") || !strcmp(argv[arg_i], "--no-trim"))
+      } else if(!strcmp(argv[arg_i], "-W") || !strcmp(argv[arg_i], "--no-trim")) {
         data.no_trim_whitespace = 1;
-      else if(!strcmp(argv[arg_i], "-d") || !strcmp(argv[arg_i], "--header-row-span")) {
-        if(!(arg_i + 1 < argc && atoi(argv[arg_i+1]) >= 0 && atoi(argv[arg_i+1]) < 256))
-          err = zsv_printerr(1, "%s option value invalid: should be integer between 1 and 255; got %s", argv[arg_i], arg_i + 1 < argc ? argv[arg_i+1] : "");
-        else
-          data.header_depth = (unsigned char)atoi(argv[++arg_i]);
       } else if(!strcmp(argv[arg_i], "--header-row")) {
         arg_i++;
         if(!(arg_i < argc))
@@ -768,12 +749,6 @@ int MAIN(int argc, const char *argv[]) {
           err = zsv_printerr(1, "%s option value invalid: should be positive integer; got %s", argv[arg_i], arg_i + 1 < argc ? argv[arg_i+1] : "");
         else
           data.data_rows_limit = atoi(argv[++arg_i]) + 1;
-      } else if(!strcmp(argv[arg_i], "-R") || !strcmp(argv[arg_i], "--skip-head")) {
-        ++arg_i;
-        if(!(arg_i < argc && atoi(argv[arg_i]) >= 0 && atoi(argv[arg_i]) < 256))
-          err = zsv_printerr(1, "-R option value invalid: should be positive integer smaller than 256");
-        else
-          data.skip_rows = data.skip_rows_orig = atoi(argv[arg_i]);
       } else if(!strcmp(argv[arg_i], "-D") || !strcmp(argv[arg_i], "--skip-data")) {
         ++arg_i;
         if(!(arg_i < argc && atoi(argv[arg_i]) >= 0))