query-forge 0.3.0

Run SQL queries on XLSX/XML/CSV/JSON/JSONL/Markdown inputs and export results as text, CSV, JSONL, Markdown, XML, XLSX, or Parquet
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
use std::path::PathBuf;

use clap::{ArgAction, ArgGroup, Args, Parser, Subcommand, ValueEnum};

#[derive(Debug, Parser)]
#[command(
    author,
    version,
    about = "Query XLSX/XML/CSV/JSON/JSONL/Markdown/Parquet datasets with SQL"
)]
pub(crate) struct Cli {
    #[command(subcommand)]
    pub(crate) command: Commands,
}

#[derive(Debug, Subcommand)]
pub(crate) enum Commands {
    /// Execute a SQL query against one or more input files.
    #[command(after_long_help = "Examples:\n\
    qf query --input ./input.xlsx:Sheet1 --sql \"SELECT * FROM table\"\n\
    qf query --input ./inventory.xml:Inventory --sql \"SELECT product, price FROM table WHERE active = 1\"\n\
    qf query --input ./inventory.csv --sql \"SELECT product, price FROM table WHERE active = 1\"\n\
    qf query --input ./inventory.json:Inventory --sql \"SELECT product, price FROM table WHERE active = 1\"\n\
    qf query --input ./inventory.jsonl --sql \"SELECT product, price FROM table WHERE active = 1\"\n\
    qf query --input ./a.xlsx:Sheet1 --input ./b.csv --sql \"SELECT * FROM table UNION ALL SELECT * FROM table2\"\n\
    qf query --input sales=./sales.xlsx:Q1 --input costs=./costs.csv --sql \"SELECT * FROM sales JOIN costs ON sales.id = costs.id\"\n\
    qf query --input ./inventory.md --sql \"SELECT product, price FROM table WHERE active = 1\"\n\
    qf query --input ./inventory.md:2 --sql \"SELECT product, price FROM table\"\n\
    qf query --input ./input.xlsx:Sheet1 --sql-file ./query.sql --output ./result.csv --format csv\n\
    qf query --input ./summary.json --sql \"SELECT key, value FROM table\" --json-mode object\n\
    qf query --input ./record.json --sql \"SELECT \\\"user.name\\\", \\\"address.city\\\" FROM table\" --json-mode flatten\n\
    qf query --input ./config.xml --sql \"SELECT tag, value FROM table WHERE tag = 'timeout'\" --xml-mode descendants\n\
    qf query --input ./items.xml --sql \"SELECT id, type, value FROM table\" --xml-mode attributes\n\
\n\
Query source:\n\
    Provide SQL inline with --sql or load it from disk with --sql-file.\n\
\n\
Table naming:\n\
    Without explicit names, inputs become table, table2, table3, ...\n\
    Prefix inputs as NAME=PATH to use readable table names in SQL.\n\
\n\
JSON extraction modes (--json-mode):\n\
    array (default): each element of a top-level array becomes a row.\n\
    object: each key-value pair of a top-level object becomes a row with 'key' and 'value' columns.\n\
    flatten: recursively flattens nested objects/arrays into a single row using dot-separated column names.\n\
\n\
XML extraction modes (--xml-mode):\n\
    rows (default): detect and extract tabular rows from the XML structure.\n\
    descendants: collect every leaf text element as a row with 'tag' and 'value' columns.\n\
    attributes: collect elements with attributes; each attribute name becomes a column.")]
    Query(QueryCommand),

    /// List logical SQL tables generated from inputs.
    #[command(after_long_help = "Examples:\n\
    qf tables --input sales=./sales.xlsx:Q1 --input costs=./costs.csv\n\
    qf tables --input ./inventory.json:Inventory --format json\n\
    qf tables --input ./inventory.xml:Inventory --input ./inventory.csv --format markdown\n\
Output:\n\
    Lists the logical SQL table names created from your inputs, in input order.\n\
    Use this command before writing queries when you want to confirm automatic table names.")]
    Tables(TablesCommand),

    /// Show per-table column names and inferred SQL-compatible types.
    #[command(after_long_help = "Examples:\n\
    qf schema --input ./inventory.json:Inventory\n\
    qf schema --input sales=./sales.xlsx:Q1 --input costs=./costs.csv --format json\n\
    qf schema --input ./inventory.xml:Inventory --format markdown\n\
Output:\n\
    Shows one row per column with the logical table name and inferred SQL-compatible type.\n\
    Use this command to verify names and types before writing filters, joins, or casts.")]
    Schema(SchemaCommand),

    /// Return a compact diagnostic summary (tables, row counts, column counts, null density, and sample rows).
    #[command(after_long_help = "Examples:\n\
    qf inspect --input ./inventory.md:2 --sample 5\n\
    qf inspect --input sales=./sales.xlsx:Q1 --input costs=./costs.csv --format json\n\
    qf inspect --input ./inventory.csv --stats --format markdown\n\
Output:\n\
    Includes a compact summary for each table, sampled rows, and optional extra metrics.\n\
    Use this command when you need fast diagnostics before or after adjusting normalization flags.")]
    Inspect(InspectCommand),
}

#[derive(Debug, Args)]
#[command(group(
    ArgGroup::new("query_source")
        .required(true)
        .args(["sql", "sql_file"])
))]
pub(crate) struct QueryCommand {
    #[arg(
        short,
        long,
        required = true,
        action = ArgAction::Append,
        value_name = "[NAME=]PATH[:SHEET|KEY]",
        long_help = "Input dataset path. Supported formats: .xlsx, .xml, .csv, .json, .jsonl, .md, .markdown, .parquet.\n\
Optionally prefix with a table name: NAME=PATH (e.g. sales=./sales.xlsx:Q1).\n\
Without a name prefix, tables are named 'table', 'table2', 'table3', ...\n\
For XLSX use PATH:SheetName (or PATH#SheetName) to select a worksheet.\n\
For XML use PATH:TagName (or PATH#TagName) to select a subtree; without sheet/tag the whole file is used.\n\
For JSON use PATH:Key (or PATH#Key) to select a top-level key; without key the whole JSON root is used.\n\
For Markdown use PATH:N (or PATH#N), where N is the 1-based table number; without key the first table is used.\n\
    CSV, JSONL and Parquet do not support sheet/key selection."
    )]
    pub(crate) input: Vec<String>,
    #[arg(
        long,
        alias = "query",
        short = 'q',
        conflicts_with = "sql_file",
        value_name = "SQL",
        long_help = "SQL query to execute. Use 'table', 'table2', 'table3', ... or explicit table names.\n\
The flag --query is also accepted as a legacy alias."
    )]
    pub(crate) sql: Option<String>,
    #[arg(
        long = "sql-file",
        value_name = "PATH",
        conflicts_with = "sql",
        long_help = "Path to a file containing the SQL query to execute.\n\
The file is read as UTF-8 text and trimmed before execution."
    )]
    pub(crate) sql_file: Option<PathBuf>,
    #[arg(
        long = "infer-types",
        conflicts_with = "all_text",
        long_help = "Infer typed SQLite values from input data when possible.\n\
This is useful for numeric comparisons, boolean filtering, and date-aware ingestion."
    )]
    pub(crate) infer_types: bool,
    #[arg(
        long = "all-text",
        conflicts_with = "infer_types",
        long_help = "Load all incoming values as text instead of inferring typed SQLite values.\n\
Use this when you want exact string preservation or need to avoid automatic conversions."
    )]
    pub(crate) all_text: bool,
    #[arg(
        long = "decimal-comma",
        long_help = "Interpret commas as decimal separators during type inference.\n\
Useful for locale-formatted values such as 12,50."
    )]
    pub(crate) decimal_comma: bool,
    #[arg(
        long = "date-format",
        value_name = "STRFTIME",
        long_help = "Expected date format used during type inference, for example %d/%m/%Y."
    )]
    pub(crate) date_format: Option<String>,
    #[arg(
        long = "null-values",
        value_delimiter = ',',
        action = ArgAction::Append,
        value_name = "VALUE[,VALUE...]",
        long_help = "Additional strings to interpret as NULL during type inference.\n\
Repeat the flag or pass comma-separated values."
    )]
    pub(crate) null_values: Vec<String>,
    #[arg(
        long = "true-values",
        value_delimiter = ',',
        action = ArgAction::Append,
        value_name = "VALUE[,VALUE...]",
        long_help = "Additional strings to interpret as boolean true during type inference.\n\
Repeat the flag or pass comma-separated values."
    )]
    pub(crate) true_values: Vec<String>,
    #[arg(
        long = "false-values",
        value_delimiter = ',',
        action = ArgAction::Append,
        value_name = "VALUE[,VALUE...]",
        long_help = "Additional strings to interpret as boolean false during type inference.\n\
Repeat the flag or pass comma-separated values."
    )]
    pub(crate) false_values: Vec<String>,
    #[arg(
        long = "trim",
        long_help = "Trim surrounding whitespace from every cell before loading it."
    )]
    pub(crate) trim: bool,
    #[arg(
        long = "skip-empty-rows",
        long_help = "Discard rows whose values are entirely empty after normalization."
    )]
    pub(crate) skip_empty_rows: bool,
    #[arg(
        long = "normalize-headers",
        long_help = "Normalize header names before loading them into SQLite.\n\
Combine with --header-case and --dedupe-headers for more stable schemas."
    )]
    pub(crate) normalize_headers: bool,
    #[arg(
        long = "header-case",
        value_enum,
        requires = "normalize_headers",
        long_help = "Header naming style to apply after normalization.\n\
Requires --normalize-headers."
    )]
    pub(crate) header_case: Option<HeaderCase>,
    #[arg(
        long = "dedupe-headers",
        long_help = "Make duplicate normalized headers unique by appending numeric suffixes."
    )]
    pub(crate) dedupe_headers: bool,
    #[arg(
        long = "param",
        value_name = "NAME=VALUE",
        action = ArgAction::Append,
        long_help = "Bind a named SQL parameter.\n\
Repeat the flag for multiple parameters; values are parsed as null, booleans, integers, reals, or text."
    )]
    pub(crate) params: Vec<String>,
    #[arg(
        short,
        long,
        value_name = "PATH",
        long_help = "Write query results to a file instead of stdout.\n\
Pair with --format to choose the serialization format when the extension is ambiguous."
    )]
    pub(crate) output: Option<PathBuf>,
    #[arg(
        long,
        value_enum,
        long_help = "Output format for query results.\n\
Supported values: text, csv, json, jsonl, markdown, xlsx, xml, parquet."
    )]
    pub(crate) format: Option<OutputFormat>,
    #[arg(
        long = "no-headers",
        long_help = "Treat the first row as data instead of column headers.\n\
Generated column names follow the pattern column1, column2, ..."
    )]
    pub(crate) no_headers: bool,
    #[arg(
        long = "json-mode",
        value_enum,
        long_help = "JSON extraction strategy.\n\
'array' (default): each element of a top-level JSON array becomes a row.\n\
'object': each key-value pair of a JSON object becomes a row with 'key' and 'value' columns.\n\
'flatten': recursively flatten nested JSON objects and arrays using dotted key paths."
    )]
    pub(crate) json_mode: Option<JsonMode>,
    #[arg(
        long = "xml-mode",
        value_enum,
        long_help = "XML extraction strategy.\n\
'rows' (default): detect and extract tabular rows from the XML structure.\n\
'descendants': collect every leaf text element as a row with 'tag' and 'value' columns.\n\
'attributes': extract element attributes as columns; each element with attributes becomes a row."
    )]
    pub(crate) xml_mode: Option<XmlMode>,
    #[arg(
        long,
        long_help = "Print query execution metadata to stderr after running the query.\n\
Includes row count, output column names, input tables loaded, and execution time.\n\
Use --meta-format to choose between text (default) and json output."
    )]
    pub(crate) meta: bool,
    #[arg(
        long = "meta-format",
        value_enum,
        requires = "meta",
        long_help = "Format for metadata output printed to stderr.\n\
Supported values: text (default), json.\n\
Requires --meta."
    )]
    pub(crate) meta_format: Option<MetadataFormat>,
}

#[derive(Debug, Args)]
pub(crate) struct TablesCommand {
    #[arg(
        short,
        long,
        required = true,
        action = ArgAction::Append,
        value_name = "[NAME=]PATH[:SHEET|KEY]",
        long_help = "Input dataset path. Supported formats: .xlsx, .xml, .csv, .json, .jsonl, .md, .markdown, .parquet.\n\
Optionally prefix with a table name: NAME=PATH (e.g. sales=./sales.xlsx:Q1).\n\
Without a name prefix, tables are named 'table', 'table2', 'table3', ...\n\
For XLSX use PATH:SheetName to select a worksheet.\n\
For XML use PATH:TagName to select a subtree.\n\
For JSON use PATH:Key to select a top-level key.\n\
For Markdown use PATH:N for the 1-based table index."
    )]
    pub(crate) input: Vec<String>,
    #[arg(
        short,
        long,
        value_name = "SELECTOR",
        long_help = "Default sheet/tag/key applied to every --input that does not already specify one."
    )]
    pub(crate) sheet: Option<String>,
    #[arg(
        long,
        value_enum,
        default_value = "text",
        long_help = "Output format for table discovery results. Supported values: text, json, markdown."
    )]
    pub(crate) format: InspectionFormat,
}

#[derive(Debug, Args)]
pub(crate) struct SchemaCommand {
    #[arg(
        short,
        long,
        required = true,
        action = ArgAction::Append,
        value_name = "[NAME=]PATH[:SHEET|KEY]",
        long_help = "Input dataset path. Supported formats: .xlsx, .xml, .csv, .json, .jsonl, .md, .markdown, .parquet.\n\
Optionally prefix with a table name: NAME=PATH (e.g. sales=./sales.xlsx:Q1).\n\
Without a name prefix, tables are named 'table', 'table2', 'table3', ...\n\
For XLSX use PATH:SheetName to select a worksheet.\n\
For XML use PATH:TagName to select a subtree.\n\
For JSON use PATH:Key to select a top-level key.\n\
For Markdown use PATH:N for the 1-based table index."
    )]
    pub(crate) input: Vec<String>,
    #[arg(
        short,
        long,
        value_name = "SELECTOR",
        long_help = "Default sheet/tag/key applied to every --input that does not already specify one."
    )]
    pub(crate) sheet: Option<String>,
    #[arg(
        long,
        value_enum,
        default_value = "text",
        long_help = "Output format for schema results. Supported values: text, json, markdown."
    )]
    pub(crate) format: InspectionFormat,
}

#[derive(Debug, Args)]
pub(crate) struct InspectCommand {
    #[arg(
        short,
        long,
        required = true,
        action = ArgAction::Append,
        value_name = "[NAME=]PATH[:SHEET|KEY]",
        long_help = "Input dataset path. Supported formats: .xlsx, .xml, .csv, .json, .jsonl, .md, .markdown, .parquet.\n\
Optionally prefix with a table name: NAME=PATH (e.g. sales=./sales.xlsx:Q1).\n\
Without a name prefix, tables are named 'table', 'table2', 'table3', ...\n\
For XLSX use PATH:SheetName to select a worksheet.\n\
For XML use PATH:TagName to select a subtree.\n\
For JSON use PATH:Key to select a top-level key.\n\
For Markdown use PATH:N for the 1-based table index."
    )]
    pub(crate) input: Vec<String>,
    #[arg(
        short,
        long,
        value_name = "SELECTOR",
        long_help = "Default sheet/tag/key applied to every --input that does not already specify one."
    )]
    pub(crate) sheet: Option<String>,
    #[arg(
        long,
        value_enum,
        default_value = "text",
        long_help = "Output format for inspection results. Supported values: text, json, markdown."
    )]
    pub(crate) format: InspectionFormat,
    /// Number of sample rows to include per table (default: 5).
    #[arg(
        long,
        default_value = "5",
        long_help = "Number of sample rows to include per table in the inspection output."
    )]
    pub(crate) sample: usize,
    /// Enable additional metrics (distinct count, min/max for numeric columns).
    #[arg(
        long,
        long_help = "Include additional metrics such as distinct counts and numeric min/max values when available."
    )]
    pub(crate) stats: bool,
}

#[derive(Clone, Debug, ValueEnum)]
pub(crate) enum OutputFormat {
    Text,
    Csv,
    Json,
    Jsonl,
    Markdown,
    Xlsx,
    Xml,
    Parquet,
}

#[derive(Clone, Debug, ValueEnum)]
pub(crate) enum InspectionFormat {
    Text,
    Json,
    Markdown,
}

#[derive(Clone, Debug, ValueEnum)]
pub(crate) enum MetadataFormat {
    /// Human-readable text output (default).
    Text,
    /// Machine-readable JSON output.
    Json,
}

#[derive(Clone, Debug, ValueEnum)]
pub(crate) enum HeaderCase {
    Snake,
}

/// JSON extraction mode for `--json-mode`.
#[derive(Clone, Debug, ValueEnum)]
pub(crate) enum JsonMode {
    /// Each element of a top-level JSON array becomes a row (default).
    Array,
    /// Each key-value pair of the JSON object becomes a row with "key" and "value" columns.
    Object,
    /// Recursively flatten nested JSON into dotted key paths (e.g. address.city).
    Flatten,
}

/// XML extraction mode for `--xml-mode`.
#[derive(Clone, Debug, ValueEnum)]
pub(crate) enum XmlMode {
    /// Detect and extract tabular rows from the XML structure (default).
    Rows,
    /// Collect every leaf text element as a row with "tag" and "value" columns.
    Descendants,
    /// Extract element attributes as columns; each element with attributes becomes a row.
    Attributes,
}

#[cfg(test)]
mod tests {
    use clap::Parser;

    use super::{Cli, Commands};

    #[test]
    fn parses_configurable_type_inference_flags() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
            "--all-text",
            "--decimal-comma",
            "--date-format",
            "%d/%m/%Y",
            "--null-values",
            "NULL,N/A",
            "--true-values",
            "YES,Y",
            "--false-values",
            "NO,N",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(cmd.all_text);
        assert!(cmd.decimal_comma);
        assert_eq!(cmd.date_format.as_deref(), Some("%d/%m/%Y"));
        assert_eq!(cmd.null_values, vec!["NULL", "N/A"]);
        assert_eq!(cmd.true_values, vec!["YES", "Y"]);
        assert_eq!(cmd.false_values, vec!["NO", "N"]);
    }

    #[test]
    fn parses_input_normalization_flags() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
            "--trim",
            "--skip-empty-rows",
            "--normalize-headers",
            "--header-case",
            "snake",
            "--dedupe-headers",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(cmd.trim);
        assert!(cmd.skip_empty_rows);
        assert!(cmd.normalize_headers);
        assert!(matches!(cmd.header_case, Some(super::HeaderCase::Snake)));
        assert!(cmd.dedupe_headers);
    }

    #[test]
    fn rejects_infer_types_and_all_text_together() {
        let parsed = Cli::try_parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
            "--infer-types",
            "--all-text",
        ]);

        assert!(parsed.is_err());
    }

    #[test]
    fn parses_sql_file_flag() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql-file",
            "query.sql",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert_eq!(cmd.sql_file.as_deref(), Some(std::path::Path::new("query.sql")));
    }

    #[test]
    fn parses_json_mode_flag() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.json",
            "--sql",
            "SELECT * FROM table",
            "--json-mode",
            "flatten",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(matches!(cmd.json_mode, Some(super::JsonMode::Flatten)));
    }

    #[test]
    fn parses_xml_mode_flag() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.xml",
            "--sql",
            "SELECT * FROM table",
            "--xml-mode",
            "attributes",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(matches!(cmd.xml_mode, Some(super::XmlMode::Attributes)));
    }

    #[test]
    fn json_mode_defaults_to_none_when_not_specified() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.json",
            "--sql",
            "SELECT * FROM table",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(cmd.json_mode.is_none());
        assert!(cmd.xml_mode.is_none());
    }

    #[test]
    fn parses_meta_flag() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
            "--meta",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(cmd.meta);
        assert!(cmd.meta_format.is_none());
    }

    #[test]
    fn parses_meta_with_json_format() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
            "--meta",
            "--meta-format",
            "json",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(cmd.meta);
        assert!(matches!(cmd.meta_format, Some(super::MetadataFormat::Json)));
    }

    #[test]
    fn parses_meta_with_text_format() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
            "--meta",
            "--meta-format",
            "text",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(cmd.meta);
        assert!(matches!(cmd.meta_format, Some(super::MetadataFormat::Text)));
    }

    #[test]
    fn meta_flag_is_false_when_not_specified() {
        let cli = Cli::parse_from([
            "qf",
            "query",
            "--input",
            "data.csv",
            "--sql",
            "SELECT * FROM table",
        ]);

        let Commands::Query(cmd) = cli.command else {
            panic!("expected query command");
        };

        assert!(!cmd.meta);
        assert!(cmd.meta_format.is_none());
    }
}