diff options
Diffstat (limited to 'source/csv_reader.h')
| -rw-r--r-- | source/csv_reader.h | 183 |
1 files changed, 0 insertions, 183 deletions
diff --git a/source/csv_reader.h b/source/csv_reader.h deleted file mode 100644 index f5205bf..0000000 --- a/source/csv_reader.h +++ /dev/null | |||
| @@ -1,183 +0,0 @@ | |||
| 1 | #ifndef ENGINE_LEXER_H | ||
| 2 | #define ENGINE_LEXER_H | ||
| 3 | |||
| 4 | typedef enum csv_token_flags csv_token_flags; | ||
| 5 | enum csv_token_flags | ||
| 6 | { | ||
| 7 | START_FL = 1 << 1, | ||
| 8 | END_FL = 1 << 2, | ||
| 9 | }; | ||
| 10 | |||
| 11 | typedef enum csv_token_type csv_token_type; | ||
| 12 | enum csv_token_type | ||
| 13 | { | ||
| 14 | // first 255 tokens for ascii characters | ||
| 15 | TOKEN_UNDEFINED = 255, | ||
| 16 | TOKEN_IDENTIFIER, | ||
| 17 | TOKEN_VALUE, | ||
| 18 | }; | ||
| 19 | |||
| 20 | typedef struct csv_token csv_token; | ||
| 21 | struct csv_token | ||
| 22 | { | ||
| 23 | string8 lexeme; | ||
| 24 | csv_token_type type; | ||
| 25 | csv_token_flags flags; | ||
| 26 | csv_token *next; | ||
| 27 | }; | ||
| 28 | |||
| 29 | // NOTE(nasr): i dont think im going to use this. | ||
| 30 | typedef struct csv_row csv_row; | ||
| 31 | struct csv_row | ||
| 32 | { | ||
| 33 | // array of size col_count, points into mmap buffer | ||
| 34 | string8 *fields; | ||
| 35 | s32 count; | ||
| 36 | }; | ||
| 37 | |||
| 38 | typedef struct csv_table csv_table; | ||
| 39 | struct csv_table | ||
| 40 | { | ||
| 41 | // first row, col names | ||
| 42 | // all data rows | ||
| 43 | string8 *headers; | ||
| 44 | csv_row *rows; | ||
| 45 | s32 col_count; | ||
| 46 | s32 row_count; | ||
| 47 | }; | ||
| 48 | |||
| 49 | |||
| 50 | typedef struct csv_token_list csv_token_list; | ||
| 51 | struct csv_token_list | ||
| 52 | { | ||
| 53 | csv_token *start_token; | ||
| 54 | csv_token *end_token; | ||
| 55 | |||
| 56 | }; | ||
| 57 | |||
| 58 | read_only global_variable | ||
| 59 | csv_token nil_csv_token= | ||
| 60 | { | ||
| 61 | .lexeme = {.data = NULL, .size =0}, | ||
| 62 | .type = (csv_token_type)0, | ||
| 63 | .flags = 0, | ||
| 64 | .next = &nil_csv_token, | ||
| 65 | |||
| 66 | }; | ||
| 67 | |||
| 68 | read_only global_variable | ||
| 69 | csv_token_list nil_csv_token_list = | ||
| 70 | { | ||
| 71 | .start_token = &nil_csv_token, | ||
| 72 | .end_token = &nil_csv_token, | ||
| 73 | }; | ||
| 74 | |||
| 75 | |||
| 76 | read_only global_variable | ||
| 77 | csv_row nil_csv_row = | ||
| 78 | { | ||
| 79 | .fields = &nil_string, | ||
| 80 | .count = 0, | ||
| 81 | }; | ||
| 82 | |||
| 83 | read_only global_variable | ||
| 84 | csv_table nil_csv_table = | ||
| 85 | { | ||
| 86 | .headers = &nil_string, | ||
| 87 | .rows = &nil_csv_row, | ||
| 88 | .col_count = 0, | ||
| 89 | .row_count = 0, | ||
| 90 | }; | ||
| 91 | |||
| 92 | #endif /* ENGINE_LEXER_H */ | ||
| 93 | |||
| 94 | // the lexer acts as a table builder from a csv file | ||
| 95 | // and parsing indivudal rows and columns | ||
| 96 | // the next step would be building a the b-tree | ||
| 97 | internal csv_token * | ||
| 98 | tokenize_csv(string8 buffer, mem_arena *arena) | ||
| 99 | { | ||
| 100 | b32 FL = TRUE; | ||
| 101 | |||
| 102 | if(buffer.size < 0) return NULL; | ||
| 103 | |||
| 104 | csv_token *tok = PushStruct(arena, csv_token); | ||
| 105 | |||
| 106 | // URGENT(nasr): segfaulting because memcpy of strring value doesnt work dammit | ||
| 107 | // NOPE ITS BEECAUSE WEE DONT LOAD CSV OR SOMTHING??? | ||
| 108 | for(s32 index = 0; buffer.data[index] != '\0'; ++index) | ||
| 109 | { | ||
| 110 | u8 point = buffer.data[index]; | ||
| 111 | |||
| 112 | s32 start = 0; | ||
| 113 | s32 end = 0; | ||
| 114 | |||
| 115 | if(is_whitespace(point)) | ||
| 116 | { | ||
| 117 | warn("csv file is invalid, detected whitespace"); | ||
| 118 | return NULL; | ||
| 119 | } | ||
| 120 | |||
| 121 | switch(point) | ||
| 122 | { | ||
| 123 | case('\n'): | ||
| 124 | { | ||
| 125 | if(FL) tok->flags |= END_FL; | ||
| 126 | break; | ||
| 127 | } | ||
| 128 | |||
| 129 | case(','): | ||
| 130 | { | ||
| 131 | end = index - 1; | ||
| 132 | start = index + 1; | ||
| 133 | break; | ||
| 134 | } | ||
| 135 | default: | ||
| 136 | { | ||
| 137 | break; | ||
| 138 | } | ||
| 139 | } | ||
| 140 | |||
| 141 | tok->lexeme = StringCast(&buffer.data[start], end - start); | ||
| 142 | tok->next = tok; | ||
| 143 | } | ||
| 144 | |||
| 145 | return tok; | ||
| 146 | } | ||
| 147 | |||
| 148 | internal void | ||
| 149 | read_csv(string8 buffer) | ||
| 150 | { | ||
| 151 | // printf("\nsize:%lu\ndata %s\n", buffer.size, buffer.data); | ||
| 152 | |||
| 153 | } | ||
| 154 | |||
| 155 | internal b_tree * | ||
| 156 | parse_csv(mem_arena *arena, csv_token_list *ctl) | ||
| 157 | { | ||
| 158 | b_tree *tree = PushStructZero(arena, b_tree); | ||
| 159 | b_tree_create(arena, tree); | ||
| 160 | |||
| 161 | //- TODO(nasr): check initizalization or something tomorrow | ||
| 162 | { | ||
| 163 | |||
| 164 | } | ||
| 165 | // TODO(nasr): fix this logic tomorrow | ||
| 166 | csv_token *ct = PushStruct(arena, csv_token); | ||
| 167 | |||
| 168 | for (;ct != NULL; ct = ct->next) | ||
| 169 | { | ||
| 170 | // skip structural ctens, only index values | ||
| 171 | if (ct->type != TOKEN_VALUE) | ||
| 172 | { | ||
| 173 | continue; | ||
| 174 | } | ||
| 175 | |||
| 176 | // NOTE(nasr): payload is the cten itself so the caller can reach | ||
| 177 | // row/col metadata without us having to copy it | ||
| 178 | // NOTE(nasr): heh why do we void cast again? | ||
| 179 | b_tree_insert(arena, tree, ct->lexeme, (void *)ct); | ||
| 180 | } | ||
| 181 | |||
| 182 | return tree; | ||
| 183 | } | ||
