2017-18 Flawless

• G-001 SCHEMA CANNOT EXPRESS 'SERIAL NUMBERED, RUN VARIES PER CARD' (EXTRACTION_SPEC 2.3), so SEVEN parallel tiers are written print_run=UNKNOWN / serial_numbered=UNKNOWN even though every one is PROVABLY serial numbered. The held facts, from complete row-level enumerations of each tier's own distinct token set (never a sample): Dual Diamond Memorabilia Gold {SN10 x19, SN8 x1 on DD-DG Dwyane Wade / Goran Dragic} https://www.tcdb.com/ViewSet.cfm/sid/189335/2017-18-Panini-Flawless-Dual-Diamond-Memorabilia-Gold ; Flawless Patch Autographs Gold {SN10 x28, SN9 x1 on card 7 Dominique Wilkins} https://www.tcdb.com/ViewSet.cfm/sid/189361/2017-18-Panini-Flawless-Flawless-Patch-Autographs-Gold ; Horizontal Patch Auto Gold {SN10 x45, SN7 x1 on HP-VC Vince Carter} https://www.tcdb.com/ViewSet.cfm/sid/189371/2017-18-Panini-Flawless-Horizontal-Patch-Auto-Gold ; Patches Ruby {SN15 on most, SN12 on P-DSH Dennis Schroder, P-DWD Dwyane Wade and P-JSL John Salley} https://www.tcdb.com/ViewSet.cfm/sid/189386/2017-18-Panini-Flawless-Patches-Ruby ; Star Swatch Signatures Ruby {SN15 x22, SN13 x1 on SS-LK Luke Kennard} https://www.tcdb.com/ViewSet.cfm/sid/189411/2017-18-Panini-Flawless-Star-Swatch-Signatures-Ruby ; Vertical Patch Auto Gold {SN10 x43, SN7 x1 on the row numbered 35, Vince Carter} https://www.tcdb.com/ViewSet.cfm/sid/189418/2017-18-Panini-Flawless-Vertical-Patch-Auto-Gold ; Vertical Patch Auto Ruby {SN15 x42, SN12 x1 on VP-JT Jayson Tatum} https://www.tcdb.com/ViewSet.cfm/sid/189421/2017-18-Panini-Flawless-Vertical-Patch-Auto-Ruby . In five of the seven a SINGLE card breaks the tier, which is the shape flawless-bkb-2018-19 also measured; a three-row sample would have written a clean sourced integer on every one of them. THE RATE: 7 of 116 tiers = 6.0%, against 6 of 79 = 7.6% on flawless-bkb-2018-19. But the COMPOSITION differs sharply from the sibling season, where FIVE of six variable tiers were Ruby: here FOUR of the seven are Gold and three are Ruby, and every one of the 24 Platinum, 8 Emerald, 4 Sapphire, 16 Printing Plates and 18 Green tiers is uniform. The 2017-18 scope memo 7.5 rule 'variability lives in the MIDDLE of each ladder; the 1/1 end never varies' holds exactly on this parent: nothing at SN1 varies, and everything that varies sits at SN10 or SN15.
• G-002 BASE SUBSET TOKENS RECORDED VERBATIM AND NOT EXPANDED. tcdb renders four contiguous subset tokens on the SN5 half of the base run - "TD" on cards 101-115, "TOC" on 116-125, "MV" on 126-140 and "CHAMP" on 141-160 - reproduced independently on the Platinum parallel sid 189324, whose own rows give the same block sizes (TD x15, TOC x10, MV x15, CHAMP x20 = 60). cardboardconnection separately names four base subsets: "The base set includes subsets: Triple Double, Top of the Class, Most Valuable, and Champions". The mapping token-to-name is obvious from initials and block order and is therefore a DERIVATION, not a source, so it is not written: base_cards.csv card_tier carries only the two sourced tier names Base Gems and Base SP Gems. UNRESOLVED: whether a source states the token-to-subset-name mapping, and whether card_tier should carry the four subsets instead of the two print-run tiers. Sources: https://www.tcdb.com/ViewSet.cfm/sid/182810/2017-18-Panini-Flawless and https://www.cardboardconnection.com/2017-18-panini-flawless-basketball-nba-cards
• G-003 SET-LEVEL PRINT RUNS THAT sets.csv CANNOT HOLD (EXTRACTION_SPEC 2.2 - there is no print_run column and card_count must not carry one). Recorded here as figures so nothing is lost. UNIFORM on a complete row-level enumeration of every rendered row: Championship Tags Autographs /2 (37 rows); Distinguished Autos /25 (40); Enshrined Signatures /25 (43); Excellence Signatures /25 (35); Flawless Autographs /25 (78); Flawless Cuts 1/1 (2); Honored Numbers /25 (55); Logoman 1/1 (25); Logoman Autographs 1/1 (54); Momentous Autographs /25 (60); Premium Ink /25 (45); Rookie Autographs /25 (15); USA Basketball /25 (12); and each of the six prior-season blocks, which are one card each - 2015-16 Update II Autographs /25, Momentous Autographed Memorabilia /25, Premium Ink /25, Star Swatch Signatures /25, 2016-17 Update I Greats /25, 2016-17 Update I Signature Prime Materials /18. Listing: https://www.tcdb.com/Inserts.cfm/sid/182810/2017-18-Panini-Flawless . Editorial comparison: https://www.cardboardconnection.com/2017-18-panini-flawless-basketball-nba-cards
• G-004 SETS WHOSE OWN ROWS DISAGREE ON THE SERIAL, so no set-level print run exists to record even in prose: Championship Tags {SN2 x35, SN1 x5}; Dual Diamond Memorabilia {SN25 x18, SN19 on DD-JC}; Dual Patches {SN25 on most plus SN20 on DP-DGL, DP-JWT and DP-TRL, SN17 on DP-ITH, SN22 on DP-JSL, SN21 on DP-RLW}; Flawless Patch Autographs {SN25 on most plus SN18 FP-DF, SN17 FP-DS, SN20 FP-JS, SN17 FP-LK}; Horizontal Patch Auto {SN25 on most plus SN22 HP-KLV, SN21 HP-ZL}; Patches {SN25 on most plus eleven named exceptions SN20 P-DGL, SN24 P-DJN, SN18 P-DWD, SN13 P-EHY, SN13 P-DSH, SN16 P-ITH, SN22 P-JSK, SN18 P-JSL, SN9 P-MTH, SN22 P-RHP, SN16 P-RLW}; Rookie Patches {SN25 on most, SN10 on RP-MMK}; Signature Prime Materials {SN25 on most, SN18 on SM-DF}; Star Swatch Signatures {SN25 x15 plus SN17 SS-JH, SN14 SS-JO, SN20 SS-JS, SN18 SS-KM}; Vertical Patch Auto {SN25 on most, SN21 on VP-KB Kobe Bryant}. Listing: https://www.tcdb.com/Inserts.cfm/sid/182810/2017-18-Panini-Flawless
• G-005 "Flawless Update" (tcdb sid 189326) IS UNRESOLVED ON THREE AXES and is written with card_count UNKNOWN. (1) COUNT: the page states "Total Cards: 31" and THREE differently-parameterised fetches each rendered exactly ONE row - card 33, DeMarcus Cousins, Sacramento Kings, no flag tokens - ViewSet.cfm, Checklist.cfm?sort=0 and Gallery.cfm?PageIndex=1. A 31-versus-1 gap is not a unit question, so neither figure is writeable. (2) SET TYPE: insert rests only on that one row carrying neither AU nor MEM; if the other thirty are autographs the row is misclassified. (3) SEASON ATTRIBUTION: its tcdb Notes cell reads "Additional cards from Panini Flawless sets that were not distributed during the original release" and names NO season, whereas the equivalent rows under the 2018-19 parent do name one. It is filed as a 2017-18 set on the strength of its sid sitting inside the 2017-18 block (189322-189421) and the absence of a season prefix in its name. Sources: https://www.tcdb.com/ViewSet.cfm/sid/189326/2017-18-Panini-Flawless-Flawless-Update and https://www.tcdb.com/Checklist.cfm/sid/189326/2017-18-Panini-Flawless-Flawless-Update?sort=0 and https://www.tcdb.com/Gallery.cfm/sid/189326/2017-18-Panini-Flawless-Flawless-Update?PageIndex=1
• G-006 COUNT DISAGREEMENTS BETWEEN tcdb AND THE EDITORIAL LINEAGE, recorded not resolved. The rendered-row count is written every time. Dual Diamond Memorabilia: cardboardconnection "20 cards" against tcdb 19 rendered rows over a header of 19. Flawless Patch Autographs Gold: cc "30 cards" against 29. Signature Prime Materials Gold: cc "37 cards" against 34. Star Swatch Signatures Gold: cc "28 cards" against 27. Vertical Patch Auto Green: cc "48 cards" against 49 - the only one where tcdb renders MORE than the editorial figure. Note also that on this product cc and checklistcenter look like ONE lineage in the CAV-037 sense: checklistcenter reproduces cc's base structure and parallel ladder word for word and supplies no independent subset counts at all, so no editorial tiebreaker was available for any of the five. https://www.cardboardconnection.com/2017-18-panini-flawless-basketball-nba-cards and https://www.checklistcenter.com/2017-18-panini-flawless-basketball/
• G-007 EDITORIAL PRINT RUNS REFUTED BY THE PRODUCT'S OWN ROWS - two unhedged figures and three ceilings. REFUTED: cardboardconnection's "Vertical Patch Auto Set Checklist: 35 cards. Serial numbered #/25." is contradicted by VP-KB (Kobe Bryant) at SN21; its "Vertical Patch Auto Ruby Set Checklist: 43 cards. Serial numbered #/15." is contradicted by VP-JT (Jayson Tatum) at SN12. CEILINGS, which are never print runs: "Flawless Patch Autographs Gold ... #/10 or as noted", "Signature Prime Materials Gold ... #/10 or as noted", "Star Swatch Signatures Gold ... #/10 or as noted". Of the three ceilinged tiers, ONE is genuinely variable (Flawless Patch Autographs Gold, SN9 on card 7) and TWO are uniform at SN10 on a complete sweep - so on this product the "or as noted" wording is 1-for-3 as a variability predictor, against 7-for-7 on flawless-bkb-2018-19's "#/25 or as Noted". The wording is a warning, not a diagnosis. https://www.cardboardconnection.com/2017-18-panini-flawless-basketball-nba-cards
• G-008 GRANULARITY DIVERGENCE FROM THE SIBLING SEASON ON PRIOR-SEASON BLOCKS, recorded rather than smoothed over. This release files ONE sets.csv row per tcdb-named member set, reproducing tcdb's verbatim season-prefixed name and a sourced card_count of 1 - six rows: "2015-16 Panini Flawless Update II: Autographs", "... Momentous Autographed Memorabilia", "... Premium Ink", "... Star Swatch Signatures", "2016-17 Panini Flawless Update I: Greats" and "... Signature Prime Materials" - plus their 24 colour parallels at set scope in parallels.csv. data/flawless-bkb-2018-19/sets.csv instead collapses each block into ONE row named after the block only ("2016-17 Panini Flawless Update II", "2017-18 Panini Flawless Update", "2017-18 Panini Flawless Update I"), each with card_count UNKNOWN and no parallels.csv rows, describing its 5 and 52 tcdb entries in prose. Both satisfy EXTRACTION_SPEC 5.5's requirement that the block be listed and mechanically filterable; they are not interchangeable for a consumer counting sets. A repair pass should pick one shape for the whole line. Listing: https://www.tcdb.com/Inserts.cfm/sid/182810/2017-18-Panini-Flawless
• G-009 SECTION 5.5 MIRROR, CHECKED IN BOTH DIRECTIONS AND CONFIRMED LIVE ON THIS LINE. Downward: this parent's Inserts.cfm files 30 rows of PRIOR-season sets - four 2015-16 Panini Flawless Update II sets and two 2016-17 Panini Flawless Update I sets, each with four colour parallels - all filed and flagged, none gapped away. Upward: our own data/flawless-bkb-2018-19/sets.csv records tcdb filing "2017-18 Panini Flawless Update" (sid 255520) and a 52-entry "2017-18 Panini Flawless Update I" block (sids 605035-605039 and 605549-605595) under the 2018-19 PARENT, so some of this release's own late cards live under a later season's parent and are NOT in this directory. That did not cost this build any base-run completeness, because the 2017-18 base run is contiguous 1-160 with no unissued interior number for an Update block to be hiding - which is the one thing the mirror check is for. A third naming exists and matches neither: checklistcenter files this product's update material as "2016 Flawless Extras Set", a string neither tcdb nor cardboardconnection uses. https://www.tcdb.com/Inserts.cfm/sid/182810/2017-18-Panini-Flawless and https://www.checklistcenter.com/2017-18-panini-flawless-basketball/
• G-010 FAMILIES IN pipeline/fix_schema.py IS STRUCTURALLY THIN FOR THIS LINE AND 63 OF 116 ROWS (54%) LAND ON 'other'. parallel_family was derived by AST-lifting FAMILIES and family_of from fix_schema.py so the repo keeps one definition, and NO entry was added (caveats.csv CAV-039: 268 hand-written parallel_family values already exist corpus-wide and re-deriving them is a known destructive trap). The list has Platinum, Emerald, Sapphire, Bronze and Printing Plate, which cover 53 rows here. It has no Ruby, no Gold, no Green, and no Logoman - so every Ruby (28 rows), Gold (23), Green (18) and the Dual Diamond Memorabilia Logoman tier resolve to 'other', as do the base-run Double and Triple. Note the list DOES contain a 'Pink' pattern mapped deliberately to 'other' to stop it matching inside 'Ink' - so 'other' here is a mix of genuinely-unclassified and deliberately-suppressed values. Ruby and Gold are the two largest single missing patterns and Gold is where this product's variability concentrates (four of seven variable tiers), so the gap is not cosmetic. NOT ACTED ON: adding patterns is outside a release build's scope.
• G-011 CHANNEL IS UNKNOWN ON 115 OF 116 PARALLEL ROWS, with channel_basis deliberately EMPTY on all 115 (migrations/003_parallels.sql enforces (channel = 'UNKNOWN') = (channel_basis IS NULL), and 'unsupported' describes a known channel that could not be sourced, not an unknown channel). The one exception is the base-run Bronze tier, where both tcdb's Inserts.cfm Notes cell ("1st Off the Line box exclusive; Serial numbered to 2. #1-100 only") and checklistcenter ("Bronze #/2 (First Off The Line Exclusive)") state the channel, giving channel=FOTL with basis sourced-exclusive. No source consulted states a trade channel, box SKU or region for any other tier, and the product-level fact that Flawless is a hobby product was NOT inherited onto tiers it does not cover. https://www.tcdb.com/Inserts.cfm/sid/182810/2017-18-Panini-Flawless and https://www.checklistcenter.com/2017-18-panini-flawless-basketball/
• G-012 CHASE-PASS FINDINGS VISIBLE FROM THIS BUILD, recorded so the insert_cards.csv pass does not rediscover them, and written NOWHERE as values in this release. (1) BARE ORDINALS mixed into letter-coded sets - the EXTRACTION_SPEC mixed-code shape: Flawless Autographs renders "16" for Gordon Hayward among FA- codes; Signature Prime Materials and its Gold, Green, Platinum and Ruby tiers render "9" for Jrue Holiday and "25" for Tim Hardaway Jr. among SM- codes; Vertical Patch Auto Platinum renders "28" Hakeem Olajuwon, "35" Vince Carter, "42" Jermaine O'Neal and "48" Lance Stephenson among VP- codes, and Gold and Ruby render "35" and "42" likewise. Per the one-real-code-settles-the-scheme rule these sets are coded and the ordinals are placeholders. (2) A LITERAL TBD CARD NUMBER: all four Vertical Patch Auto Printing Plates tiers render the string TBD in the card-number column for Lance Stephenson, confirmed character for character on two endpoints. It is unwritable as a card_number and the player is named here so the roster survives.
• G-013 Errors.cfm ON THE PARENT RETURNED NOTHING AND THAT IS NOT EVIDENCE (caveats.csv CAV-020). https://www.tcdb.com/Errors.cfm/sid/182810/2017-18-Panini-Flawless is headed "Errors / Variations" and says "No errors or variations have been added for this set yet." - the identical sentence the endpoint returns on sets that provably DO carry UER rows, so it discriminates nothing. NO CLAIM IS MADE that this release has no variation set. What the base run does rest on is an independent contiguous enumeration: 160 rendered rows over 160 distinct numbers, with Gallery.cfm pages 8, 9, 10, 13 and 16 each rendering card numbers equal to their own index range, which proves no number below 160 is unissued and leaves no room for a b-suffixed variation layer inside the base numbering. A variation set filed as duplicate-numbered b rows would not be surfaced by that enumeration and remains unsupported rather than disproved. https://www.tcdb.com/ViewSet.cfm/sid/182810/2017-18-Panini-Flawless
• G-014 SOURCE BREADTH. Beckett was not attempted: the 2017-18 scope memo records it confirmed unreachable for this season (302 to a maintenance page). No marketplace corroboration was sought, because every print run written here rests on a complete row-level enumeration of a tcdb tier page rather than on an editorial integer, and no plain integer was written onto an autograph set from an aggregator. cardboardconnection and checklistcenter behave as ONE lineage on this product - checklistcenter reproduces cc's base structure and parallel ladder essentially word for word and carries no independent subset counts - so no checkable divergence between them was found here, unlike on Cornerstones. That is why source_confidence is 'single' on 105 of 116 parallel rows: tcdb's row sweep is one leg and the editorial lineage is at most one more. https://www.cardboardconnection.com/2017-18-panini-flawless-basketball-nba-cards and https://www.checklistcenter.com/2017-18-panini-flawless-basketball/

If you know a source that answers one of these, tell us — name the question's label and where to look.