gitea

Development moved to Codeberg

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
  59. 59
  60. 60
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67
  68. 68
  69. 69
  70. 70
  71. 71
  72. 72
  73. 73
  74. 74
  75. 75
  76. 76
  77. 77
  78. 78
  79. 79
  80. 80
  81. 81
  82. 82
  83. 83
  84. 84
  85. 85
  86. 86
  87. 87
  88. 88
  89. 89
  90. 90
  91. 91
  92. 92
  93. 93
  94. 94
  95. 95
  96. 96
  97. 97
  98. 98
  99. 99
  100. 100
  101. 101
  102. 102
  103. 103
  104. 104
  105. 105
  106. 106
  107. 107
  108. 108
  109. 109
  110. 110
  111. 111
  112. 112
  113. 113
  114. 114
  115. 115
  116. 116
  117. 117
  118. 118
  119. 119
  120. 120
  121. 121
  122. 122
  123. 123
  124. 124
  125. 125
  126. 126
  127. 127
  128. 128
  129. 129
  130. 130
  131. 131
  132. 132
  133. 133
  134. 134
  135. 135
  136. 136
  137. 137
  138. 138
  139. 139
  140. 140
  141. 141
  142. 142
  143. 143
  144. 144
  145. 145
  146. 146
  147. 147
  148. 148
  149. 149
  150. 150
  151. 151
  152. 152
  153. 153
  154. 154
  155. 155
  156. 156
  157. 157
  158. 158
  159. 159
  160. 160
  161. 161
  162. 162
  163. 163
  164. 164
  165. 165
  166. 166
  167. 167
  168. 168
  169. 169
  170. 170
  171. 171
  172. 172
  173. 173
  174. 174
  175. 175
  176. 176
  177. 177
  178. 178
  179. 179
  180. 180
  181. 181
  182. 182
  183. 183
  184. 184
  185. 185
  186. 186
  187. 187
  188. 188
  189. 189
  190. 190
  191. 191
  192. 192
  193. 193
  194. 194
  195. 195
  196. 196
  197. 197
  198. 198
  199. 199
  200. 200
  201. 201
  202. 202
  203. 203
  204. 204
  205. 205
  206. 206
  207. 207
  208. 208
  209. 209
  210. 210
  211. 211
  212. 212
  213. 213
  214. 214
  215. 215
  216. 216
  217. 217
  218. 218
  219. 219
  220. 220
  221. 221
  222. 222
  223. 223
  224. 224
  225. 225
  226. 226
  227. 227
  228. 228
  229. 229
  230. 230
  231. 231
  232. 232
  233. 233
  234. 234
  235. 235
  236. 236
  237. 237
  238. 238
  239. 239
  240. 240
  241. 241
  242. 242
  243. 243
  244. 244
  245. 245
  246. 246
  247. 247
  248. 248
  249. 249
  250. 250
  251. 251
  252. 252
  253. 253
  254. 254
  255. 255
  256. 256
  257. 257
  258. 258
  259. 259
  260. 260
  261. 261
  262. 262
  263. 263
  264. 264
  265. 265
  266. 266
  267. 267
  268. 268
  269. 269
  270. 270
  271. 271
  272. 272
  273. 273
  274. 274
  275. 275
  276. 276
  277. 277
  278. 278
  279. 279
  280. 280
  281. 281
  282. 282
  283. 283
  284. 284
  285. 285
  286. 286
  287. 287
  288. 288
  289. 289
  290. 290
  291. 291
  292. 292
  293. 293
  294. 294
  295. 295
  296. 296
  297. 297
  298. 298
  299. 299
  300. 300
  301. 301
  302. 302
  303. 303
  304. 304
  305. 305
  306. 306
  307. 307
  308. 308
  309. 309
  310. 310
  311. 311
  312. 312
  313. 313
  314. 314
  315. 315
  316. 316
  317. 317
  318. 318
  319. 319
  320. 320
  321. 321
  322. 322
  323. 323
  324. 324
  325. 325
  326. 326
  327. 327
  328. 328
  329. 329
  330. 330
  331. 331
  332. 332
  333. 333
  334. 334
  335. 335
  336. 336
  337. 337
  338. 338
  339. 339
  340. 340
  341. 341
  342. 342
  343. 343
  344. 344
  345. 345
  346. 346
  347. 347
  348. 348
  349. 349
  350. 350
  351. 351
  352. 352
  353. 353
  354. 354
  355. 355
  356. 356
  357. 357
  358. 358
  359. 359
  360. 360
  361. 361
  362. 362
  363. 363
  364. 364
  365. 365
  366. 366
  367. 367
  368. 368
  369. 369
  370. 370
  371. 371
  372. 372
  373. 373
  374. 374
  375. 375
  376. 376
  377. 377
  378. 378
  379. 379
  380. 380
  381. 381
  382. 382
  383. 383
  384. 384
  385. 385
  386. 386
  387. 387
  388. 388
  389. 389
  390. 390
  391. 391
  392. 392
  393. 393
  394. 394
  395. 395
  396. 396
  397. 397
  398. 398
  399. 399
  400. 400
  401. 401
  402. 402
  403. 403
  404. 404
  405. 405
  406. 406
  407. 407
  408. 408
  409. 409
  410. 410
  411. 411
  412. 412
  413. 413
  414. 414
  415. 415
  416. 416
  417. 417
  418. 418
  419. 419
  420. 420
  421. 421
  422. 422
  423. 423
  424. 424
  425. 425
  426. 426
  427. 427
  428. 428
  429. 429
  430. 430
  431. 431
  432. 432
  433. 433
  434. 434
  435. 435
  436. 436
  437. 437
  438. 438
  439. 439
  440. 440
  441. 441
  442. 442
  443. 443
  444. 444
  445. 445
  446. 446
  447. 447
  448. 448
  449. 449
  450. 450
  451. 451
  452. 452
  453. 453
  454. 454
  455. 455
  456. 456
  457. 457
  458. 458
  459. 459
  460. 460
  461. 461
  462. 462
  463. 463
  464. 464
  465. 465
  466. 466
  467. 467
  468. 468
  469. 469
  470. 470
  471. 471
  472. 472
  473. 473
  474. 474
  475. 475
  476. 476
  477. 477
  478. 478
  479. 479
  480. 480
  481. 481
  482. 482
  483. 483
  484. 484
  485. 485
  486. 486
  487. 487
  488. 488
  489. 489
  490. 490
  491. 491
  492. 492
  493. 493
  494. 494
  495. 495
  496. 496
  497. 497
  498. 498
  499. 499
  500. 500
  501. 501
  502. 502
  503. 503
  504. 504
  505. 505
  506. 506
  507. 507
  508. 508
  509. 509
  510. 510
  511. 511
  512. 512
  513. 513
  514. 514
  515. 515
  516. 516
  517. 517
  518. 518
  519. 519
  520. 520
  521. 521
  522. 522
  523. 523
  524. 524
  525. 525
  526. 526
  527. 527
  528. 528
  529. 529
  530. 530
  531. 531
  532. 532
  533. 533
  534. 534
  535. 535
  536. 536
  537. 537
  538. 538
  539. 539
  540. 540
  541. 541
  542. 542
  543. 543
  544. 544
  545. 545
  546. 546
  547. 547
  548. 548
  549. 549
  550. 550
  551. 551
  552. 552
  553. 553
  554. 554
  555. 555
  556. 556
  557. 557
  558. 558
  559. 559
  560. 560
  561. 561
  562. 562
  563. 563
  564. 564
  565. 565
  566. 566
  567. 567
  568. 568
  569. 569
  570. 570
  571. 571
  572. 572
  573. 573
  574. 574
  575. 575
  576. 576
  577. 577
  578. 578
  579. 579
  580. 580
  581. 581
  582. 582
  583. 583
  584. 584
  585. 585
  586. 586
  587. 587
  588. 588
  589. 589
  590. 590
  591. 591
  592. 592
  593. 593
  594. 594
  595. 595
  596. 596
  597. 597
  598. 598
  599. 599
  600. 600
  601. 601
  602. 602
  603. 603
  604. 604
  605. 605
  606. 606
  607. 607
  608. 608
  609. 609
  610. 610
  611. 611
  612. 612
  613. 613
  614. 614
  615. 615
  616. 616
  617. 617
  618. 618
  619. 619
  620. 620
  621. 621
  622. 622
  623. 623
  624. 624
  625. 625
  626. 626
  627. 627
  628. 628
  629. 629
  630. 630
  631. 631
  632. 632
  633. 633
  634. 634
  635. 635
  636. 636
  637. 637
  638. 638
  639. 639
  640. 640
  641. 641
  642. 642
  643. 643
  644. 644
  645. 645
  646. 646
  647. 647
  648. 648
  649. 649
  650. 650
  651. 651
  652. 652
  653. 653
  654. 654
  655. 655
  656. 656
  657. 657
  658. 658
  659. 659
  660. 660
  661. 661
  662. 662
  663. 663
  664. 664
  665. 665
  666. 666
  667. 667
  668. 668
  669. 669
  670. 670
  671. 671
  672. 672
  673. 673
  674. 674
  675. 675
  676. 676
  677. 677
  678. 678
  679. 679
  680. 680
  681. 681
  682. 682
  683. 683
  684. 684
  685. 685
  686. 686
  687. 687
  688. 688
  689. 689
  690. 690
  691. 691
  692. 692
  693. 693
  694. 694
  695. 695
  696. 696
  697. 697
  698. 698
  699. 699
  700. 700
  701. 701
  702. 702
  703. 703
  704. 704
  705. 705
  706. 706
  707. 707
  708. 708
  709. 709
  710. 710
  711. 711
  712. 712
  713. 713
  714. 714
  715. 715
  716. 716
  717. 717
  718. 718
  719. 719
  720. 720
  721. 721
  722. 722
  723. 723
  724. 724
  725. 725
  726. 726
  727. 727
  728. 728
  729. 729
  730. 730
  731. 731
  732. 732
  733. 733
  734. 734
  735. 735
  736. 736
  737. 737
  738. 738
  739. 739
  740. 740
  741. 741
  742. 742
  743. 743
  744. 744
  745. 745
  746. 746
  747. 747
  748. 748
  749. 749
  750. 750
  751. 751
  752. 752
  753. 753
  754. 754
  755. 755
  756. 756
  757. 757
  758. 758
  759. 759
  760. 760
  761. 761
  762. 762
  763. 763
  764. 764
  765. 765
  766. 766
  767. 767
  768. 768
  769. 769
  770. 770
  771. 771
  772. 772
  773. 773
  774. 774
  775. 775
// Copyright 2014 The Prometheus Authors
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

package expfmt

import (
	"bufio"
	"bytes"
	"fmt"
	"io"
	"math"
	"strconv"
	"strings"

	dto "github.com/prometheus/client_model/go"

	"github.com/golang/protobuf/proto"
	"github.com/prometheus/common/model"
)

// A stateFn is a function that represents a state in a state machine. By
// executing it, the state is progressed to the next state. The stateFn returns
// another stateFn, which represents the new state. The end state is represented
// by nil.
type stateFn func() stateFn

// ParseError signals errors while parsing the simple and flat text-based
// exchange format.
type ParseError struct {
	Line int
	Msg  string
}

// Error implements the error interface.
func (e ParseError) Error() string {
	return fmt.Sprintf("text format parsing error in line %d: %s", e.Line, e.Msg)
}

// TextParser is used to parse the simple and flat text-based exchange format. Its
// zero value is ready to use.
type TextParser struct {
	metricFamiliesByName map[string]*dto.MetricFamily
	buf                  *bufio.Reader // Where the parsed input is read through.
	err                  error         // Most recent error.
	lineCount            int           // Tracks the line count for error messages.
	currentByte          byte          // The most recent byte read.
	currentToken         bytes.Buffer  // Re-used each time a token has to be gathered from multiple bytes.
	currentMF            *dto.MetricFamily
	currentMetric        *dto.Metric
	currentLabelPair     *dto.LabelPair

	// The remaining member variables are only used for summaries/histograms.
	currentLabels map[string]string // All labels including '__name__' but excluding 'quantile'/'le'
	// Summary specific.
	summaries       map[uint64]*dto.Metric // Key is created with LabelsToSignature.
	currentQuantile float64
	// Histogram specific.
	histograms    map[uint64]*dto.Metric // Key is created with LabelsToSignature.
	currentBucket float64
	// These tell us if the currently processed line ends on '_count' or
	// '_sum' respectively and belong to a summary/histogram, representing the sample
	// count and sum of that summary/histogram.
	currentIsSummaryCount, currentIsSummarySum     bool
	currentIsHistogramCount, currentIsHistogramSum bool
}

// TextToMetricFamilies reads 'in' as the simple and flat text-based exchange
// format and creates MetricFamily proto messages. It returns the MetricFamily
// proto messages in a map where the metric names are the keys, along with any
// error encountered.
//
// If the input contains duplicate metrics (i.e. lines with the same metric name
// and exactly the same label set), the resulting MetricFamily will contain
// duplicate Metric proto messages. Similar is true for duplicate label
// names. Checks for duplicates have to be performed separately, if required.
// Also note that neither the metrics within each MetricFamily are sorted nor
// the label pairs within each Metric. Sorting is not required for the most
// frequent use of this method, which is sample ingestion in the Prometheus
// server. However, for presentation purposes, you might want to sort the
// metrics, and in some cases, you must sort the labels, e.g. for consumption by
// the metric family injection hook of the Prometheus registry.
//
// Summaries and histograms are rather special beasts. You would probably not
// use them in the simple text format anyway. This method can deal with
// summaries and histograms if they are presented in exactly the way the
// text.Create function creates them.
//
// This method must not be called concurrently. If you want to parse different
// input concurrently, instantiate a separate Parser for each goroutine.
func (p *TextParser) TextToMetricFamilies(in io.Reader) (map[string]*dto.MetricFamily, error) {
	p.reset(in)
	for nextState := p.startOfLine; nextState != nil; nextState = nextState() {
		// Magic happens here...
	}
	// Get rid of empty metric families.
	for k, mf := range p.metricFamiliesByName {
		if len(mf.GetMetric()) == 0 {
			delete(p.metricFamiliesByName, k)
		}
	}
	// If p.err is io.EOF now, we have run into a premature end of the input
	// stream. Turn this error into something nicer and more
	// meaningful. (io.EOF is often used as a signal for the legitimate end
	// of an input stream.)
	if p.err == io.EOF {
		p.parseError("unexpected end of input stream")
	}
	return p.metricFamiliesByName, p.err
}

func (p *TextParser) reset(in io.Reader) {
	p.metricFamiliesByName = map[string]*dto.MetricFamily{}
	if p.buf == nil {
		p.buf = bufio.NewReader(in)
	} else {
		p.buf.Reset(in)
	}
	p.err = nil
	p.lineCount = 0
	if p.summaries == nil || len(p.summaries) > 0 {
		p.summaries = map[uint64]*dto.Metric{}
	}
	if p.histograms == nil || len(p.histograms) > 0 {
		p.histograms = map[uint64]*dto.Metric{}
	}
	p.currentQuantile = math.NaN()
	p.currentBucket = math.NaN()
}

// startOfLine represents the state where the next byte read from p.buf is the
// start of a line (or whitespace leading up to it).
func (p *TextParser) startOfLine() stateFn {
	p.lineCount++
	if p.skipBlankTab(); p.err != nil {
		// End of input reached. This is the only case where
		// that is not an error but a signal that we are done.
		p.err = nil
		return nil
	}
	switch p.currentByte {
	case '#':
		return p.startComment
	case '\n':
		return p.startOfLine // Empty line, start the next one.
	}
	return p.readingMetricName
}

// startComment represents the state where the next byte read from p.buf is the
// start of a comment (or whitespace leading up to it).
func (p *TextParser) startComment() stateFn {
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentByte == '\n' {
		return p.startOfLine
	}
	if p.readTokenUntilWhitespace(); p.err != nil {
		return nil // Unexpected end of input.
	}
	// If we have hit the end of line already, there is nothing left
	// to do. This is not considered a syntax error.
	if p.currentByte == '\n' {
		return p.startOfLine
	}
	keyword := p.currentToken.String()
	if keyword != "HELP" && keyword != "TYPE" {
		// Generic comment, ignore by fast forwarding to end of line.
		for p.currentByte != '\n' {
			if p.currentByte, p.err = p.buf.ReadByte(); p.err != nil {
				return nil // Unexpected end of input.
			}
		}
		return p.startOfLine
	}
	// There is something. Next has to be a metric name.
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.readTokenAsMetricName(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentByte == '\n' {
		// At the end of the line already.
		// Again, this is not considered a syntax error.
		return p.startOfLine
	}
	if !isBlankOrTab(p.currentByte) {
		p.parseError("invalid metric name in comment")
		return nil
	}
	p.setOrCreateCurrentMF()
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentByte == '\n' {
		// At the end of the line already.
		// Again, this is not considered a syntax error.
		return p.startOfLine
	}
	switch keyword {
	case "HELP":
		return p.readingHelp
	case "TYPE":
		return p.readingType
	}
	panic(fmt.Sprintf("code error: unexpected keyword %q", keyword))
}

// readingMetricName represents the state where the last byte read (now in
// p.currentByte) is the first byte of a metric name.
func (p *TextParser) readingMetricName() stateFn {
	if p.readTokenAsMetricName(); p.err != nil {
		return nil
	}
	if p.currentToken.Len() == 0 {
		p.parseError("invalid metric name")
		return nil
	}
	p.setOrCreateCurrentMF()
	// Now is the time to fix the type if it hasn't happened yet.
	if p.currentMF.Type == nil {
		p.currentMF.Type = dto.MetricType_UNTYPED.Enum()
	}
	p.currentMetric = &dto.Metric{}
	// Do not append the newly created currentMetric to
	// currentMF.Metric right now. First wait if this is a summary,
	// and the metric exists already, which we can only know after
	// having read all the labels.
	if p.skipBlankTabIfCurrentBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	return p.readingLabels
}

// readingLabels represents the state where the last byte read (now in
// p.currentByte) is either the first byte of the label set (i.e. a '{'), or the
// first byte of the value (otherwise).
func (p *TextParser) readingLabels() stateFn {
	// Summaries/histograms are special. We have to reset the
	// currentLabels map, currentQuantile and currentBucket before starting to
	// read labels.
	if p.currentMF.GetType() == dto.MetricType_SUMMARY || p.currentMF.GetType() == dto.MetricType_HISTOGRAM {
		p.currentLabels = map[string]string{}
		p.currentLabels[string(model.MetricNameLabel)] = p.currentMF.GetName()
		p.currentQuantile = math.NaN()
		p.currentBucket = math.NaN()
	}
	if p.currentByte != '{' {
		return p.readingValue
	}
	return p.startLabelName
}

// startLabelName represents the state where the next byte read from p.buf is
// the start of a label name (or whitespace leading up to it).
func (p *TextParser) startLabelName() stateFn {
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentByte == '}' {
		if p.skipBlankTab(); p.err != nil {
			return nil // Unexpected end of input.
		}
		return p.readingValue
	}
	if p.readTokenAsLabelName(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentToken.Len() == 0 {
		p.parseError(fmt.Sprintf("invalid label name for metric %q", p.currentMF.GetName()))
		return nil
	}
	p.currentLabelPair = &dto.LabelPair{Name: proto.String(p.currentToken.String())}
	if p.currentLabelPair.GetName() == string(model.MetricNameLabel) {
		p.parseError(fmt.Sprintf("label name %q is reserved", model.MetricNameLabel))
		return nil
	}
	// Special summary/histogram treatment. Don't add 'quantile' and 'le'
	// labels to 'real' labels.
	if !(p.currentMF.GetType() == dto.MetricType_SUMMARY && p.currentLabelPair.GetName() == model.QuantileLabel) &&
		!(p.currentMF.GetType() == dto.MetricType_HISTOGRAM && p.currentLabelPair.GetName() == model.BucketLabel) {
		p.currentMetric.Label = append(p.currentMetric.Label, p.currentLabelPair)
	}
	if p.skipBlankTabIfCurrentBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentByte != '=' {
		p.parseError(fmt.Sprintf("expected '=' after label name, found %q", p.currentByte))
		return nil
	}
	// Check for duplicate label names.
	labels := make(map[string]struct{})
	for _, l := range p.currentMetric.Label {
		lName := l.GetName()
		if _, exists := labels[lName]; !exists {
			labels[lName] = struct{}{}
		} else {
			p.parseError(fmt.Sprintf("duplicate label names for metric %q", p.currentMF.GetName()))
			return nil
		}
	}
	return p.startLabelValue
}

// startLabelValue represents the state where the next byte read from p.buf is
// the start of a (quoted) label value (or whitespace leading up to it).
func (p *TextParser) startLabelValue() stateFn {
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentByte != '"' {
		p.parseError(fmt.Sprintf("expected '\"' at start of label value, found %q", p.currentByte))
		return nil
	}
	if p.readTokenAsLabelValue(); p.err != nil {
		return nil
	}
	if !model.LabelValue(p.currentToken.String()).IsValid() {
		p.parseError(fmt.Sprintf("invalid label value %q", p.currentToken.String()))
		return nil
	}
	p.currentLabelPair.Value = proto.String(p.currentToken.String())
	// Special treatment of summaries:
	// - Quantile labels are special, will result in dto.Quantile later.
	// - Other labels have to be added to currentLabels for signature calculation.
	if p.currentMF.GetType() == dto.MetricType_SUMMARY {
		if p.currentLabelPair.GetName() == model.QuantileLabel {
			if p.currentQuantile, p.err = parseFloat(p.currentLabelPair.GetValue()); p.err != nil {
				// Create a more helpful error message.
				p.parseError(fmt.Sprintf("expected float as value for 'quantile' label, got %q", p.currentLabelPair.GetValue()))
				return nil
			}
		} else {
			p.currentLabels[p.currentLabelPair.GetName()] = p.currentLabelPair.GetValue()
		}
	}
	// Similar special treatment of histograms.
	if p.currentMF.GetType() == dto.MetricType_HISTOGRAM {
		if p.currentLabelPair.GetName() == model.BucketLabel {
			if p.currentBucket, p.err = parseFloat(p.currentLabelPair.GetValue()); p.err != nil {
				// Create a more helpful error message.
				p.parseError(fmt.Sprintf("expected float as value for 'le' label, got %q", p.currentLabelPair.GetValue()))
				return nil
			}
		} else {
			p.currentLabels[p.currentLabelPair.GetName()] = p.currentLabelPair.GetValue()
		}
	}
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	switch p.currentByte {
	case ',':
		return p.startLabelName

	case '}':
		if p.skipBlankTab(); p.err != nil {
			return nil // Unexpected end of input.
		}
		return p.readingValue
	default:
		p.parseError(fmt.Sprintf("unexpected end of label value %q", p.currentLabelPair.GetValue()))
		return nil
	}
}

// readingValue represents the state where the last byte read (now in
// p.currentByte) is the first byte of the sample value (i.e. a float).
func (p *TextParser) readingValue() stateFn {
	// When we are here, we have read all the labels, so for the
	// special case of a summary/histogram, we can finally find out
	// if the metric already exists.
	if p.currentMF.GetType() == dto.MetricType_SUMMARY {
		signature := model.LabelsToSignature(p.currentLabels)
		if summary := p.summaries[signature]; summary != nil {
			p.currentMetric = summary
		} else {
			p.summaries[signature] = p.currentMetric
			p.currentMF.Metric = append(p.currentMF.Metric, p.currentMetric)
		}
	} else if p.currentMF.GetType() == dto.MetricType_HISTOGRAM {
		signature := model.LabelsToSignature(p.currentLabels)
		if histogram := p.histograms[signature]; histogram != nil {
			p.currentMetric = histogram
		} else {
			p.histograms[signature] = p.currentMetric
			p.currentMF.Metric = append(p.currentMF.Metric, p.currentMetric)
		}
	} else {
		p.currentMF.Metric = append(p.currentMF.Metric, p.currentMetric)
	}
	if p.readTokenUntilWhitespace(); p.err != nil {
		return nil // Unexpected end of input.
	}
	value, err := parseFloat(p.currentToken.String())
	if err != nil {
		// Create a more helpful error message.
		p.parseError(fmt.Sprintf("expected float as value, got %q", p.currentToken.String()))
		return nil
	}
	switch p.currentMF.GetType() {
	case dto.MetricType_COUNTER:
		p.currentMetric.Counter = &dto.Counter{Value: proto.Float64(value)}
	case dto.MetricType_GAUGE:
		p.currentMetric.Gauge = &dto.Gauge{Value: proto.Float64(value)}
	case dto.MetricType_UNTYPED:
		p.currentMetric.Untyped = &dto.Untyped{Value: proto.Float64(value)}
	case dto.MetricType_SUMMARY:
		// *sigh*
		if p.currentMetric.Summary == nil {
			p.currentMetric.Summary = &dto.Summary{}
		}
		switch {
		case p.currentIsSummaryCount:
			p.currentMetric.Summary.SampleCount = proto.Uint64(uint64(value))
		case p.currentIsSummarySum:
			p.currentMetric.Summary.SampleSum = proto.Float64(value)
		case !math.IsNaN(p.currentQuantile):
			p.currentMetric.Summary.Quantile = append(
				p.currentMetric.Summary.Quantile,
				&dto.Quantile{
					Quantile: proto.Float64(p.currentQuantile),
					Value:    proto.Float64(value),
				},
			)
		}
	case dto.MetricType_HISTOGRAM:
		// *sigh*
		if p.currentMetric.Histogram == nil {
			p.currentMetric.Histogram = &dto.Histogram{}
		}
		switch {
		case p.currentIsHistogramCount:
			p.currentMetric.Histogram.SampleCount = proto.Uint64(uint64(value))
		case p.currentIsHistogramSum:
			p.currentMetric.Histogram.SampleSum = proto.Float64(value)
		case !math.IsNaN(p.currentBucket):
			p.currentMetric.Histogram.Bucket = append(
				p.currentMetric.Histogram.Bucket,
				&dto.Bucket{
					UpperBound:      proto.Float64(p.currentBucket),
					CumulativeCount: proto.Uint64(uint64(value)),
				},
			)
		}
	default:
		p.err = fmt.Errorf("unexpected type for metric name %q", p.currentMF.GetName())
	}
	if p.currentByte == '\n' {
		return p.startOfLine
	}
	return p.startTimestamp
}

// startTimestamp represents the state where the next byte read from p.buf is
// the start of the timestamp (or whitespace leading up to it).
func (p *TextParser) startTimestamp() stateFn {
	if p.skipBlankTab(); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.readTokenUntilWhitespace(); p.err != nil {
		return nil // Unexpected end of input.
	}
	timestamp, err := strconv.ParseInt(p.currentToken.String(), 10, 64)
	if err != nil {
		// Create a more helpful error message.
		p.parseError(fmt.Sprintf("expected integer as timestamp, got %q", p.currentToken.String()))
		return nil
	}
	p.currentMetric.TimestampMs = proto.Int64(timestamp)
	if p.readTokenUntilNewline(false); p.err != nil {
		return nil // Unexpected end of input.
	}
	if p.currentToken.Len() > 0 {
		p.parseError(fmt.Sprintf("spurious string after timestamp: %q", p.currentToken.String()))
		return nil
	}
	return p.startOfLine
}

// readingHelp represents the state where the last byte read (now in
// p.currentByte) is the first byte of the docstring after 'HELP'.
func (p *TextParser) readingHelp() stateFn {
	if p.currentMF.Help != nil {
		p.parseError(fmt.Sprintf("second HELP line for metric name %q", p.currentMF.GetName()))
		return nil
	}
	// Rest of line is the docstring.
	if p.readTokenUntilNewline(true); p.err != nil {
		return nil // Unexpected end of input.
	}
	p.currentMF.Help = proto.String(p.currentToken.String())
	return p.startOfLine
}

// readingType represents the state where the last byte read (now in
// p.currentByte) is the first byte of the type hint after 'HELP'.
func (p *TextParser) readingType() stateFn {
	if p.currentMF.Type != nil {
		p.parseError(fmt.Sprintf("second TYPE line for metric name %q, or TYPE reported after samples", p.currentMF.GetName()))
		return nil
	}
	// Rest of line is the type.
	if p.readTokenUntilNewline(false); p.err != nil {
		return nil // Unexpected end of input.
	}
	metricType, ok := dto.MetricType_value[strings.ToUpper(p.currentToken.String())]
	if !ok {
		p.parseError(fmt.Sprintf("unknown metric type %q", p.currentToken.String()))
		return nil
	}
	p.currentMF.Type = dto.MetricType(metricType).Enum()
	return p.startOfLine
}

// parseError sets p.err to a ParseError at the current line with the given
// message.
func (p *TextParser) parseError(msg string) {
	p.err = ParseError{
		Line: p.lineCount,
		Msg:  msg,
	}
}

// skipBlankTab reads (and discards) bytes from p.buf until it encounters a byte
// that is neither ' ' nor '\t'. That byte is left in p.currentByte.
func (p *TextParser) skipBlankTab() {
	for {
		if p.currentByte, p.err = p.buf.ReadByte(); p.err != nil || !isBlankOrTab(p.currentByte) {
			return
		}
	}
}

// skipBlankTabIfCurrentBlankTab works exactly as skipBlankTab but doesn't do
// anything if p.currentByte is neither ' ' nor '\t'.
func (p *TextParser) skipBlankTabIfCurrentBlankTab() {
	if isBlankOrTab(p.currentByte) {
		p.skipBlankTab()
	}
}

// readTokenUntilWhitespace copies bytes from p.buf into p.currentToken.  The
// first byte considered is the byte already read (now in p.currentByte).  The
// first whitespace byte encountered is still copied into p.currentByte, but not
// into p.currentToken.
func (p *TextParser) readTokenUntilWhitespace() {
	p.currentToken.Reset()
	for p.err == nil && !isBlankOrTab(p.currentByte) && p.currentByte != '\n' {
		p.currentToken.WriteByte(p.currentByte)
		p.currentByte, p.err = p.buf.ReadByte()
	}
}

// readTokenUntilNewline copies bytes from p.buf into p.currentToken.  The first
// byte considered is the byte already read (now in p.currentByte).  The first
// newline byte encountered is still copied into p.currentByte, but not into
// p.currentToken. If recognizeEscapeSequence is true, two escape sequences are
// recognized: '\\' translates into '\', and '\n' into a line-feed character.
// All other escape sequences are invalid and cause an error.
func (p *TextParser) readTokenUntilNewline(recognizeEscapeSequence bool) {
	p.currentToken.Reset()
	escaped := false
	for p.err == nil {
		if recognizeEscapeSequence && escaped {
			switch p.currentByte {
			case '\\':
				p.currentToken.WriteByte(p.currentByte)
			case 'n':
				p.currentToken.WriteByte('\n')
			default:
				p.parseError(fmt.Sprintf("invalid escape sequence '\\%c'", p.currentByte))
				return
			}
			escaped = false
		} else {
			switch p.currentByte {
			case '\n':
				return
			case '\\':
				escaped = true
			default:
				p.currentToken.WriteByte(p.currentByte)
			}
		}
		p.currentByte, p.err = p.buf.ReadByte()
	}
}

// readTokenAsMetricName copies a metric name from p.buf into p.currentToken.
// The first byte considered is the byte already read (now in p.currentByte).
// The first byte not part of a metric name is still copied into p.currentByte,
// but not into p.currentToken.
func (p *TextParser) readTokenAsMetricName() {
	p.currentToken.Reset()
	if !isValidMetricNameStart(p.currentByte) {
		return
	}
	for {
		p.currentToken.WriteByte(p.currentByte)
		p.currentByte, p.err = p.buf.ReadByte()
		if p.err != nil || !isValidMetricNameContinuation(p.currentByte) {
			return
		}
	}
}

// readTokenAsLabelName copies a label name from p.buf into p.currentToken.
// The first byte considered is the byte already read (now in p.currentByte).
// The first byte not part of a label name is still copied into p.currentByte,
// but not into p.currentToken.
func (p *TextParser) readTokenAsLabelName() {
	p.currentToken.Reset()
	if !isValidLabelNameStart(p.currentByte) {
		return
	}
	for {
		p.currentToken.WriteByte(p.currentByte)
		p.currentByte, p.err = p.buf.ReadByte()
		if p.err != nil || !isValidLabelNameContinuation(p.currentByte) {
			return
		}
	}
}

// readTokenAsLabelValue copies a label value from p.buf into p.currentToken.
// In contrast to the other 'readTokenAs...' functions, which start with the
// last read byte in p.currentByte, this method ignores p.currentByte and starts
// with reading a new byte from p.buf. The first byte not part of a label value
// is still copied into p.currentByte, but not into p.currentToken.
func (p *TextParser) readTokenAsLabelValue() {
	p.currentToken.Reset()
	escaped := false
	for {
		if p.currentByte, p.err = p.buf.ReadByte(); p.err != nil {
			return
		}
		if escaped {
			switch p.currentByte {
			case '"', '\\':
				p.currentToken.WriteByte(p.currentByte)
			case 'n':
				p.currentToken.WriteByte('\n')
			default:
				p.parseError(fmt.Sprintf("invalid escape sequence '\\%c'", p.currentByte))
				return
			}
			escaped = false
			continue
		}
		switch p.currentByte {
		case '"':
			return
		case '\n':
			p.parseError(fmt.Sprintf("label value %q contains unescaped new-line", p.currentToken.String()))
			return
		case '\\':
			escaped = true
		default:
			p.currentToken.WriteByte(p.currentByte)
		}
	}
}

func (p *TextParser) setOrCreateCurrentMF() {
	p.currentIsSummaryCount = false
	p.currentIsSummarySum = false
	p.currentIsHistogramCount = false
	p.currentIsHistogramSum = false
	name := p.currentToken.String()
	if p.currentMF = p.metricFamiliesByName[name]; p.currentMF != nil {
		return
	}
	// Try out if this is a _sum or _count for a summary/histogram.
	summaryName := summaryMetricName(name)
	if p.currentMF = p.metricFamiliesByName[summaryName]; p.currentMF != nil {
		if p.currentMF.GetType() == dto.MetricType_SUMMARY {
			if isCount(name) {
				p.currentIsSummaryCount = true
			}
			if isSum(name) {
				p.currentIsSummarySum = true
			}
			return
		}
	}
	histogramName := histogramMetricName(name)
	if p.currentMF = p.metricFamiliesByName[histogramName]; p.currentMF != nil {
		if p.currentMF.GetType() == dto.MetricType_HISTOGRAM {
			if isCount(name) {
				p.currentIsHistogramCount = true
			}
			if isSum(name) {
				p.currentIsHistogramSum = true
			}
			return
		}
	}
	p.currentMF = &dto.MetricFamily{Name: proto.String(name)}
	p.metricFamiliesByName[name] = p.currentMF
}

func isValidLabelNameStart(b byte) bool {
	return (b >= 'a' && b <= 'z') || (b >= 'A' && b <= 'Z') || b == '_'
}

func isValidLabelNameContinuation(b byte) bool {
	return isValidLabelNameStart(b) || (b >= '0' && b <= '9')
}

func isValidMetricNameStart(b byte) bool {
	return isValidLabelNameStart(b) || b == ':'
}

func isValidMetricNameContinuation(b byte) bool {
	return isValidLabelNameContinuation(b) || b == ':'
}

func isBlankOrTab(b byte) bool {
	return b == ' ' || b == '\t'
}

func isCount(name string) bool {
	return len(name) > 6 && name[len(name)-6:] == "_count"
}

func isSum(name string) bool {
	return len(name) > 4 && name[len(name)-4:] == "_sum"
}

func isBucket(name string) bool {
	return len(name) > 7 && name[len(name)-7:] == "_bucket"
}

func summaryMetricName(name string) string {
	switch {
	case isCount(name):
		return name[:len(name)-6]
	case isSum(name):
		return name[:len(name)-4]
	default:
		return name
	}
}

func histogramMetricName(name string) string {
	switch {
	case isCount(name):
		return name[:len(name)-6]
	case isSum(name):
		return name[:len(name)-4]
	case isBucket(name):
		return name[:len(name)-7]
	default:
		return name
	}
}

func parseFloat(s string) (float64, error) {
	if strings.ContainsAny(s, "pP_") {
		return 0, fmt.Errorf("unsupported character in float")
	}
	return strconv.ParseFloat(s, 64)
}