2018-19 Donruss Optic Premium Box Set

• G-001 BASE-CARD IDENTITY COLLISION between TWO releases, per EXTRACTION_SPEC 5.2. This release (opticpremium-bkb-2018-19) and data/optic-bkb-2018-19 describe THE SAME 200 physical base card slots. Measured, not estimated: all 200 rows were read independently off this product's own tcdb sid 197339 and then diffed positionally in code against data/optic-bkb-2018-19/base_cards.csv. THE OVERLAP IS 200 OF 200 ON CARD NUMBERS, 200 OF 200 ON PLAYER NAMES AND 200 OF 200 ON TEAMS - a complete overlap with zero differences on any field. The Premium Box Set is therefore the FULL parent base run repackaged, not a subset of it, and the 201 in cardboardconnection's "201-card factory set" is 200 base plus 1 autograph, not 201 base slots. Built as instructed and NOT deduplicated and NOT double-counted; the duplication is deliberate and must be resolved at catalogue level, not by deleting either file. This is the seventh consecutive season of this line to measure a complete overlap and the earliest one in the catalogue. Sources: https://www.tcdb.com/Checklist.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set?PageIndex=2&sort=0 ; https://www.tcdb.com/Gallery.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set?PageIndex=20 ; https://www.tcdb.com/ViewSet.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set
• G-002 THE OPEN QUESTION THIS BUILD WAS COMMISSIONED TO RESOLVE, AND THE ANSWER: the 2018-19 object IS a genuine separately-sold product, and this release was therefore written. It resolves gap G-007 of data/optic-bkb-2018-19, which met the object, wrote no row for it, and recorded the decision as needing a call. THE FACTS THAT SETTLE IT, none of which a full base parallel can have. ONE, its own box and its own sale channel: cardboardconnection runs a dedicated product page, H1 "2018-19 Donruss Optic Premium Box Set Basketball Cards", "Product Configuration:" "201-card factory set", "Direct from the Panini online store, 2018-19 Donruss Optic Premium Box Set Basketball offers a complete set", box contents "Full base set (200 cards)" and "1 Rated Rookie Autograph" (https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards). TWO, its own release date, unqualified - see G-004. THREE, AND DECISIVELY, CONTENT THE PARENT'S BASE RUN DOES NOT HAVE: a 25-card autograph set exclusive to this format. tcdb gives it its own sid, 197356, "2018-19 Donruss Optic - Rated Rookies Signatures Premium Box Set", "Total Cards: 25", every row flagged "AU, SN10" (https://www.tcdb.com/ViewSet.cfm/sid/197356/2018-19-Donruss-Optic-Rated-Rookies-Signatures-Premium-Box-Set), and it is a different sid from the parent's own Rated Rookies Signatures, 197342. checklistcenter annotates that set "Notes: Exclusive to Premium Factory Set Format" and "25 Cards - Serial Numbered #/10" (https://www.checklistcenter.com/2018-19-donruss-optic-basketball/). cardboardconnection: "Collectors can also chase the 25 different Rated Rookie autograph cards, including Doncic. Numbered to only 10, these come one per box." FOUR, checklistcenter describes the object as a format, not a tier - "Separate Premium Factory Box Version is Serial numbered #/249" - and its base-set parallel ladder for the parent ("Holo (Retail); Blue Velocity (Multi-Pack); Checkerboard; Hyper Pink; Purple; Shock; Orange #/199; Lime Green #/149; Red #/99; Pink Velocity #/79; Blue #/49; Black Velocity #/39; Pink #/25; Purple Stars #/13 (First Off the Line); Gold #/10; Green #/5; Black 1/1; gold Vinyl 1/1") does NOT contain a Premium Box Set line. WHAT IS TRUE ON THE OTHER SIDE, recorded so this is not read as a one-sided case: the complete 200-of-200 overlap at G-001 means the base checklist is structurally indistinguishable from a full base parallel, and tcdb genuinely does file the base half of this product as a row inside the parent's "Parallel Sets (31)" section - named "Premium Box Set", sid 197339, Notes "Serial numbered to 249" (https://www.tcdb.com/Inserts.cfm/sid/191579/2018-19-Donruss-Optic) - and files no separate product for it on the season index: https://www.tcdb.com/ViewAll.cfm/sp/Basketball/year/2018 lists no 2018 basketball product whose name contains "Premium", and exactly two Optic entries, "2018-19 Donruss Optic" sid 191579 and "2018-19 Panini Contenders Optic" sid 206591. The overlap alone was therefore treated as NOT settling the question, in either direction; the autograph set is what settles it. See also G-003 and G-012.
• G-003 parallels.csv is deliberately EMPTY, header only, zero data rows, per EXTRACTION_SPEC 2.3 - and the absence is EVIDENCED, not unresolved. No parallel of the Premium Box Set base run was found on any source. tcdb's complete set inventory for the 2018-19 Optic parent was fetched in a single untruncated response - all three section headings "Insert Sets (141)", "Mail-In Sets (1)" and "Parallel Sets (31)" and every row under each - and it contains EXACTLY TWO rows whose name contains "Premium": "Premium Box Set" sid 197339, Notes "Serial numbered to 249", and "Rated Rookies Signatures Premium Box Set" sid 197356, Notes "4.0% of images added" (https://www.tcdb.com/Inserts.cfm/sid/191579/2018-19-Donruss-Optic). Those are this release's base set and its autograph set; neither is a parallel OF this release. checklistcenter carries the whole product as the single line "Separate Premium Factory Box Version is Serial numbered #/249" plus the one autograph line, with no colour tiers (https://www.checklistcenter.com/2018-19-donruss-optic-basketball/), and cardboardconnection's product page carries no parallel list at all (https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards). CONTRAST WITH A LATER SEASON, so this is not mistaken for a coverage hole: data/opticpremium-bkb-2024-25/parallels.csv carries three rows - Blue #/15, Gold #/7, Green #/3 - all scoped to the AUTOGRAPH set, not to the base run. No such tiers are catalogued for 2018-19 on any source fetched. data/opticpremium-bkb-2019-20 likewise writes an empty parallels.csv, its G-004.
• G-004 release_date 2019-02-08 rests on ONE source, and that is the only weakness in it. cardboardconnection prints the label "Release Date" followed by "February 8, 2019" (https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards). The label was checked specifically against EXTRACTION_SPEC 3.2's estimated-date rule: a narrow re-fetch under a distinct URL was asked to write out the entire label that immediately precedes the date, character for character, and to answer YES or NO whether it contains the word "Estimated"; it returned the label "Release Date:" and answered NO. The date is therefore written rather than gapped to UNKNOWN. NOTE THIS IS A REVERSAL OF THE SIBLING SEASONS AND A GAIN, NOT A REGRESSION: opticpremium-bkb-2019-20 G-003 records that its only date was printed as "Estimated Release Date: March 24, 2020", and opticpremium-bkb-2024-25 G-003 records "Estimated Release Date:" September 11, 2025; both correctly wrote UNKNOWN. The earliest season of the line is the one with an unqualified date. Corroboration was sought and not obtained: tcdb carries no release date on sid 197339 (https://www.tcdb.com/ViewSet.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set), checklistcenter publishes no date for the format (https://www.checklistcenter.com/2018-19-donruss-optic-basketball/), and beckett.com is unreachable - see G-013.
• G-005 base_set_size is written as 200 while the most quotable headline figure on the only dedicated product page is 201: cardboardconnection "Product Configuration:" "201-card factory set" (https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards). 200 was written because the SAME page decomposes the 201 itself, as box contents "Full base set (200 cards)" plus "1 Rated Rookie Autograph"; because tcdb's header for sid 197339 reads "Total Cards: 200" over cards 1-200 (https://www.tcdb.com/ViewSet.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set); and because Gallery page 20 of 20 ends at card 200 and the navigation offers no page 21 (https://www.tcdb.com/Gallery.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set?PageIndex=20). EXTRACTION_SPEC 5.3 requires the autograph to be excluded. Flagged because any later source quoting "201 cards" will look like a coverage hole against this file and is not one. The identical hazard is recorded at opticpremium-bkb-2019-20 G-011 and opticpremium-bkb-2024-25 G-009.
• G-006 NUMBER OF SETS OR BOXES PRINTED IS UNKNOWN AND NOTHING WAS INFERRED - the unit trap, EXTRACTION_SPEC 5.4. No source fetched states how many boxes or sets were produced. The temptation is real: cardboardconnection writes "The chromium set adds an overlapping holographic circle background to every premium stock card, which is only available in this format, plus a numbered print run of 249 copies." - a sentence in which "the chromium set" is the grammatical subject and "249" could be read as a number of SETS. It was not used for base_print_run. What base_print_run rests on instead is the unambiguously PER-CARD sentence on the same page, "Although every card is number to 249, and all 200 cards come in each set, the box contains a random assortment featuring various combinations of card #1/249 to #249/249.", which distinguishes the two units itself - "every card" numbered to 249, and separately "all 200 cards come in each set" - and by the per-copy notation "#1/249 to #249/249". A set count of 249 would also not be arithmetically clean against the 25-card autograph set at #/10, which yields 250 autographs against a hypothetical 249 boxes at one per box. No set or box count is written in any column. Source: https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards. The same non-inference is recorded at opticpremium-bkb-2019-20 G-002.
• G-007 rookie_flag BASIS IS THE LITERAL TOKEN "RR", NOT AN "RC" DESIGNATION, AND THE SAME PHYSICAL CHECKLIST CARRIES DIFFERENT TOKENS UNDER THE TWO SIDS. Both were checked and both are quoted literally here, with no expansion of either token. UNDER THE BOX-SET SID 197339, rows 151-200 render "RR, SN249" and rows 1-150 render "SN249" alone - sampled at 151 Jarred Vanderbilt, 152 Jerome Robinson, 153 Melvin Frazier Jr., 160 Mo Bamba, 170 Wendell Carter Jr. (https://www.tcdb.com/Checklist.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set?PageIndex=2&sort=0 and its &sort=1 variant) and at 171-200 on Gallery pages 18, 19 and 20, against 101, 110 and 140 which carry "SN249" only. UNDER THE PARENT SID 191579 the same rows render "RR, RC" - verified directly at 151, 152, 157, 160, 162, 164, 167, 168 and 170, with cards 101, 110 and 140 carrying no flag token at all (https://www.tcdb.com/Checklist.cfm/sid/191579/2018-19-Donruss-Optic?PageIndex=2). So the box-set sid substitutes "SN249" where the parent sid prints "RC", and the string "RC" appears NOWHERE on this product's own pages. EXTRACTION_SPEC 2.4 says to write yes/no "only where the source explicitly shows an RC designation". rookie_flag=yes was written for 151-200 and no for 1-150 on the "RR" block boundary, which is identical on both sids and matches what data/optic-bkb-2018-19/base_cards.csv carries for the identical checklist, 150 no and 50 yes. If the spec means RC strictly, all 200 rows here should be UNKNOWN and both releases need the same correction; the identical conflict is recorded at opticpremium-bkb-2019-20 G-010 and opticpremium-bkb-2024-25 G-008. NO GLOSS OF "RR" IS RECORDED ANYWHERE IN THIS RELEASE - see G-015.
• G-008 The autograph set's card numbers DUPLICATE base card numbers WITHIN this release. Rated Rookies Signatures Premium Box Set renders 25 rows at numbers 152, 154, 156, 157, 160, 162, 164, 166, 167, 168, 170, 174, 175, 177, 179, 180, 182, 184, 188, 189, 190, 192, 194, 198 and 200 (https://www.tcdb.com/ViewSet.cfm/sid/197356/2018-19-Donruss-Optic-Rated-Rookies-Signatures-Premium-Box-Set), every one of which is also a base card number in base_cards.csv. Any card-number-keyed join across this release's sets will collide unless it is also keyed by set. Counted by hand off the rendered rows, 25 of them, agreeing with the page header "Total Cards: 25". Flagged now because the card-level autograph checklist belongs to a later pass under CHASE_CHECKLIST_SPEC. The same shape is recorded at opticpremium-bkb-2019-20 G-007. NOTE the signer at a shared number is not always the base card's player: base card 157 is Deandre Ayton and autograph 157 is Deandre Ayton, but this was not checked slot by slot and the 2019-20 build measured a substitution at its own card 174.
• G-009 SCHEMA EXPRESSIVENESS: the autograph set's print run cannot be held in sets.csv. Every row of Rated Rookies Signatures Premium Box Set is serial numbered to 10 - tcdb renders the literal tokens "AU, SN10" on all 25 rows (https://www.tcdb.com/ViewSet.cfm/sid/197356/2018-19-Donruss-Optic-Rated-Rookies-Signatures-Premium-Box-Set), checklistcenter states "25 Cards - Serial Numbered #/10" (https://www.checklistcenter.com/2018-19-donruss-optic-basketball/), cardboardconnection states "Numbered to only 10, these come one per box." (https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards). EXTRACTION_SPEC 2.2 is explicit that sets.csv has no print_run column, that adding one is out of scope, and that the figure must NOT go into card_count because that field means how many cards. The figure is therefore carried in that set's notes in words and here, and card_count holds 25, the count of distinct cards. It is also not written as a parallels.csv row, because the numbering is the set's own uniform property rather than a tier of it - EXTRACTION_SPEC 2.2's "Extravagance Autographs, all #/25" case exactly. This is the fourth consecutive season of this line to hit the same wall; opticpremium-bkb-2024-25 could express its /15, /7 and /3 tiers only because they were genuine parallel tiers of the autograph set.
• G-010 CONTRADICTION FOUND AGAINST EXISTING DATA, recorded and not resolved here. checklistcenter states the PARENT product's autograph set "Rated Rookies Signatures" is "44 Cards" (https://www.checklistcenter.com/2018-19-donruss-optic-basketball/), while data/optic-bkb-2018-19/sets.csv carries that same set at card_count 45. This release directory holds no row for that set and nothing here depends on the figure; it is recorded because it was seen. No file outside this release directory was touched. A repair pass should re-count tcdb sid 197342 directly rather than take either number.
• G-011 STRUCTURAL DIVERGENCE INSIDE THE CATALOGUE about WHERE the box-set autograph object lives, now live in two directories for one season. THIS release files the 25-card object as an auto SET in sets.csv, matching data/opticpremium-bkb-2019-20 ("Rated Rookies Signatures Premium Box Set", set_type auto, card_count 25) and data/opticpremium-bkb-2024-25 (same name, auto, card_count 12). data/optic-bkb-2018-19/parallels.csv independently files THE SAME tcdb sid 197356 as a PARALLEL row - parallel_name "Premium Box Set", scope "Rated Rookies Signatures", scope_type "set", print_run UNKNOWN - citing https://www.tcdb.com/ViewSet.cfm/sid/197356/2018-19-Donruss-Optic-Rated-Rookies-Signatures-Premium-Box-Set. Both filings are defensible from where each agent stood and neither is wrong about the object; the catalogue now describes one physical set twice under two release ids, in the same shape as the base-card collision at G-001. A SECOND, RELATED DIVERGENCE on the BASE half: data/optic-bkb-2019-20/parallels.csv carries a base-set-scope row "Premium Box Set Base" print_run 249 for its season's equivalent object WHILE data/opticpremium-bkb-2019-20 also exists as a release, so 2019-20 holds the base object twice; data/optic-bkb-2018-19 wrote no such row, so with this release built, 2018-19 holds it exactly once. That is the more consistent state, but it means the two adjacent seasons of the parent line now differ, and a reader comparing optic-bkb-2018-19 with optic-bkb-2019-20 will see a sourced tier missing from the earlier one. Nothing outside this directory was touched. Decide at catalogue level, not by editing either file in isolation.
• G-012 THIS LINE PREDATES THE CATALOGUE BY A FURTHER SEASON, noted and deliberately NOT built - out of scope per the task prompt. tcdb carries a 2017-18 edition at sid 171625: fetched, it renders the set name "2017-18 Donruss Optic - Premium Box Set", "Total Cards: 200" and Notes "Serial numbered to 249", with its first rows "1 DeAndre' Bembry, Atlanta Hawks, SN249", "2 Dennis Schroder, Atlanta Hawks, SN249", "3 Taurean Prince, Atlanta Hawks, SN249" (https://www.tcdb.com/ViewSet.cfm/sid/171625/2017-18-Donruss-Optic-Premium-Box-Set). So the structure is identical a season earlier - same size, same stated run, same per-row token - and the roster is ordered differently, alphabetically by team rather than by the 2018-19 ordering, which is a further sign each season's checklist is its own. Whether a 2017-18 Rated Rookies Signatures Premium Box Set autograph set exists was NOT checked and no 2017-18 aggregator page was fetched. data/ contains no 2017-18 release of any line, so building it would also need a parent that does not exist here.
• G-013 BECKETT REMAINS UNREACHABLE - tenth consecutive pass. https://www.beckett.com/news/2018-19-donruss-optic-basketball-premium-box-set/ was found by web search with the title "2018-19 Donruss Optic Basketball Premium Box Sets Launched by Panini" and, fetched, returns HTTP 302 to https://maintenance.beckett.com/. A Panini launch announcement for exactly this product therefore EXISTS and is the obvious independent corroboration for the release date at G-004 and for the sale channel, and it could not be read. EXTRACTION_SPEC 5.4's warning applies in full: with Beckett gone, cardboardconnection and checklistcenter are the only two text sources and they are known to copy each other, so where both are cited they may be one opinion. What is genuinely independent of them here is tcdb's per-row flag rendering and the Amazon marketplace listing.
• G-014 SINGLE-SOURCE AND UNRESOLVED ITEMS, listed so they are not mistaken for settled values. (a) The release_date is single-source - see G-004. (b) The base checklist's only structural source is tcdb; no aggregator publishes a card-by-card Premium Box Set checklist for 2018-19, so the ROSTER is single-source at card level even though the set size and the print run have four sources between them. (c) The autograph set name "Rated Rookies Signatures Premium Box Set" is how tcdb and checklistcenter render it; no source shows what is printed on the card, so whether "Premium Box Set" is part of the set name or a cataloguer's qualifier separating it from the parent's own 45-card Rated Rookies Signatures is unresolved - the identical question is open at opticpremium-bkb-2019-20 G-012(a). (d) scarcity_tier for the autograph set is UNKNOWN; no source states a tier, and stated_odds "1 per box" rests on cardboardconnection alone (https://www.cardboardconnection.com/2018-19-donruss-optic-premium-box-set-basketball-cards). (e) SOURCE CALIBRATION, EXTRACTION_SPEC 3.3 and check 6, was run before any of this was trusted: checklistcenter was tested on a question already answered in this repository - the parent's base parallel ladder - and its print runs Orange #/199, Lime Green #/149, Red #/99, Pink Velocity #/79, Blue #/49, Black Velocity #/39, Pink #/25, Purple Stars #/13, Gold #/10, Green #/5, Black 1/1 agree with data/optic-bkb-2018-19/parallels.csv on 11 of 11 figures (https://www.checklistcenter.com/2018-19-donruss-optic-basketball/). cardboardconnection was tested on two facts checkable against tcdb - the /249 per-card run and the 25-card autograph set at /10 - and passed both. Neither was given an uncheckable question before passing a checkable one.
• G-015 FETCH-LAYER DEFECT, recorded because it produced a false statement in this pass and EXTRACTION_SPEC 3.4 predicts exactly it. Asked to list the Gallery tiles for cards 171-180, the fetch layer returned the correct tiles and tokens and then added, unprompted, the sentence that the flags indicate "Rookie Rendition" status. That expansion of "RR" appears nowhere on the page. The task briefing records that this layer has already invented TWO contradictory expansions of "RR"; this is a third. NO GLOSS OF ANY FLAG TOKEN IS RECORDED ANYWHERE IN THIS RELEASE - "RR", "RC", "SN249", "SN10" and "AU" are carried as literal strings only, and the block boundary at 151 is what rookie_flag actually rests on, not any reading of what the letters stand for. Source of the defect: https://www.tcdb.com/Gallery.cfm/sid/197339/2018-19-Donruss-Optic-Premium-Box-Set?PageIndex=18
• G-016 LICENCE STATUS IS UNSOURCED AS TO GRANTOR, and the value was given by the task prompt rather than found. license_status "licensed (NBA)" is the catalogue-wide value and was written as instructed, with licensed=yes and league=NBA. No source fetched in this pass states a licensing grantor for this product: tcdb prints no licensing statement on sid 197339 or 197356, cardboardconnection's product page prints none, checklistcenter prints none. The four earliest seasons of this line carry "licensed (NBA/NBPA)" and the later ones "licensed (NBA)"; data/opticpremium-bkb-2019-20's release notes carry a dated NORMALIZED marker explaining that the split was never a disagreement between sources but between agents, and that the weaker claim is the correct one where the grantor is unverified. This release is written consistently with that reasoning. Nothing was promoted.
• G-017 METHOD OBSERVATIONS ON THE FETCH LAYER, measured in this pass, offered for EXTRACTION_SPEC 3.2. (a) Checklist.cfm PageIndex=1 on sid 197339 truncated at ROW 73 of 100 - inside the documented 71-to-86 band and near its floor - and the truncation reported itself as the end of the page ("content cuts off after card 73"). PageIndex=2 truncated at row 70 of its 100, i.e. at card 170. Rows 74-100 and 171-200 were recovered from Gallery.cfm pages 8, 9, 10, 18, 19 and 20 at 10 tiles per page, exactly the top-up method the spec prescribes; the real cost was 9 fetches for 200 cards against 20 for Gallery alone. (b) THE RENUMBERING CONTROL WAS RUN AND PASSED. A named-player probe was made on the pages transcribed: "Kyrie Irving" returned card 56 and "Trae Young" card 198, and both are independently corroborated - 56 by an Amazon listing title "2018-19 Donruss Optic Premium Basketball Box Set #56 Kyrie Irving 69/249", and 198 by the autograph checklist on sid 197356, which renders "198 Trae Young Atlanta Hawks". "Luka Doncic" returned 177 on Gallery page 18 and 177 again on the autograph page. No constant-shift defect was detected. (c) Gallery page 20 was fetched twice under DIFFERENT URLs, bare and with &sort=0, and returned 10 tiles and the same ten card numbers both times, so the count is two observations rather than one cache hit. (d) Cards 71, 72 and 73 were rendered by BOTH Checklist.cfm page 1 and Gallery.cfm page 8 and agree on all three rows, which ties the two transcription methods together at their seam. (e) Inserts.cfm on the 2018-19 Optic parent again did NOT truncate, returning all three section headings and every row in one response - confirming data/optic-bkb-2018-19 G-002 against the briefing's expectation, on a second independent occasion three days later.

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