Show page

Traveline Wiki

About me · Miscellaneous

====== Differences ====== This shows you the differences between two versions of the page.

Link to this comparison view

tnds [2012/06/15 21:41]
admin Issues 29-64 transferred 		tnds [2024/01/31 09:48] (current)
Line 1: Line 1:
- * Listed - most recent first *+======Traveline National Dataset====== ​ 
  
-Template - cut and paste +To find out more about TNDS and to sign up to download the data go to [http://​traveline.info/​tnds] ​
-Issue No: +
-Where detected (and date):  +
-Observation:​  +
-Hypothesis:​ +
-Raised by: +
-Assigned ​to:+
  
- +In [[TNDSissues]] you can see issues that have been raised ​with the data and how they are being dealt with.
-Issue No: 64 +
-Where detected (and date): TNDS data 3 June 2012 +
-Observation:​ a single route can have multiple +
-files +
-Hypothesis: An explanation would be beneficial ​  +
-Raised by: Alan  +
-Assigned to: Peter Stoner +
-Peter Stoner comments: The incidences of multiple files will mostly be found, though not necessarily exclusively,​ in the NW, Wales, Yorkshire and Scottish data.  The data in these areas, tends to be held with separate tables for different periods, rather than holding the changes in a single file.  The highest number of files we have found is over 100 for a single route. ​ The causes for such a high number of files are being investigated. +
- +
- +
-Issue No: 63 +
-Where detected (and date): '​SVRACAO001-20100228-217.xml'​ and other files from '​W.zip'​ 29/​05/​2012 +
-Observation:​ Non-unique JourneyPatternTimingLink ids, violates schema uniqueness constraint +
-Hypothesis: This file contains two JourneyPatternSections,​ with ids '​JPSection2'​ and '​JPSection3'​ +
-            the ids of the links in both these sections are the same, numbers 9850 to 9895. +
-            This means there 92 TimingLinks but only 46 unique IDs. +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: +
- +
-Issue No: 62 +
-Where detected (and date): File NE_Darington_Data.20 in NE.ZIP, downloaded 2012-05-29 +
-Observation:​ I noticed that this bus service has an OperatingPeriod element with StartDate of 3rd June, i.e. the future. ​ However, when calling the Nextbuses API (e.g. for StopPoint with atcoCode 07605043), it shows this service is scheduled to arrive at the stop every hour today. ​ (at xx:​54) ​ Both of these cannot be correct. +
-Hypothesis: Either the NextBuses API is producing incorrect schedules, or Bus Service with lineName '​20'​ is marked with a start date in the future, but should have a start date in the past. +
-Raised by: Carl Partridge +
-Assigned to: +
- +
-Issue No: 61 +
-Where detected (and date): ​In TNDS files downloaded on 16th May 2012, e.g. NE_Cumbria_Data_1.xml +
-Observation: ​ In some files (such as NE_Cumbria_Data_1.xml) the Service elements have no Description sub-element. ​ I haven'​t checked if this element is mandatory, but it exists in around 90% the national data.  It's obviously a very helpful element to have in the data set, since it providers a quick, user-facing description of a bus service. ​ The affected Service elements are mostly in  London, Cumbria or Durham. +
-Hypothesis: Description element of some Service elements is missing. +
-Raised by:  Carl Partridge +
-Assigned to: Peter stoner +
-Peter Stoner comments: ​ Issue 53 also concerns Service Descriptions. ​ While the field may be desirable, it is not just that it is populated, but what it is populated with. +
- +
- +
-Issue No: 60  CLOSED +
-Where detected (and date): In London files downloaded on 16th May 2012, e.g.  output_txc_54C11j.xml +
-Observation: ​ In many (if not all) of the files for the London area (such as output_txc_54C11j.xml) various sub-nodes of the Operator element appear to be abbreviated forms of what they are meant to be.  For example, '​OperatorNameOnLicence'​ is set to ML, which is unlikely to be what is on the actual licence. ​ Could these perhaps be correctly populated with the correct operator name? +
-Hypothesis: The OperatorNameOnLicence, ​ OperatorTradingName and OperatorShortName elements are abbreviated the same as the operator short code instead of their long-form equivalents. +
-Raised by:  Carl Partridge +
-Assigned to: Peter Stoner +
-Peter Stoner comments: These fields are intended for Electronic Bus Registration which will not necessarily be being used, and certainly not in London, where bus registration does not apply. ​ Better indentification of operators will come as we improve the National Operator Code dataset and get it working in all regions. +
- +
- +
- +
-Issue No: 59 +
-Where detected (and date):​set_6-26B-1-y08.xml 12/05/12 +
-Observation:​ TransXchange 2.4 reports  +
-PublishingException:​ Problem with input document: couldn'​t find RouteLinkRef element for  +
-JourneyPatternTimingLink with id: JPL_6-26-B-y08-1-UL-196-0-2 . +
-Hypothesis: Each Journey pattern link does not have an actual or inherited link to the route scheme. +
-This prevents mapping and appears to be a conversion error. If the mapping is turned off a matrix is  +
-produced which is significantly different from the actual timetable. The source of this difference  +
-is not yet obvious. +
-Raised by: Philip Mart +
-Assigned to: +
- +
-Issue No: 58 +
-Where detected (and date):​set_6-26B-1-y08.xml 12/05/12 +
-Observation:​ TransXchange 2.4 reports +
-Line 16768, column 46: Duplicate unique value [6-26-B-y08-1] declared for identity constraint of element +
-Hypothesis: The &​lt;​PrivateCode&​gt;​ child element of each &​lt;​Service&​gt;​ is not unique. When amended to be unique +
-the fault is removed. It would appear to be a conversion error. I found that the error occurs in all of  +
-the files I checked in the SE.zip +
-Raised by: Philip Mart +
-Assigned to: +
- +
-Issue No: 57 +
-Where detected (and date): SVRSKAO041-20120124-1105.xml,​ 12/05 +
-Observation:​ Having uploaded all files I find there is no route for the reverse of SKAO041_003. While the service '​hasMirror',​ this cannot be strictly true as the reverse route uses the a different set of stops (on the other side of the road). Have seen other similar examples from the WM.zip archive. +
-Hypothesis:​ +
-Raised by: Tim Millington. +
-Assigned to: Keith Sabin (WM) +
- +
-Issue No: 56 +
-Where detected (and date): WM.zip downloaded 05/05, observed 11/05 +
-Observation:​ Seems to be a lot of duplicate information. E.g., the information in SVRSKAO041-20120124-1104.xml seems to be repeated in SVRSKAO041-20120124-1105.xml.Should I be assumin that the information on the file with the higher number it the end of the file name supercedes that with the lower? +
-EDIT: I have noticed that the latter file is Weekend Operation, while the first Weekday, so there is a difference. Since both files reference the same set of JourneyPatternSections (ref issue 55, below), is it safe to assume the JourneyPatternSections information is duplicate?​ +
-Hypothesis:​ +
-Raised by: Tim Millington +
-Assigned to: Keith Sabin (WM) +
- +
-Issue No: 55 +
-Where detected (and date): WM.zip downloaded 05/05, observed 10/05 +
-Observation:​ All the files in WM.zip (e.g. SVRSKAO041-20120124-1104.xml) seem to use generic &​lt;​JourneyPatternSection&​gt;​ ids, e.g., &​quot;​JPSection2&​quot;​. According to the schema guide, this element is a top level element -- so should these ids be unique across the system? +
-Hypothesis:​ +
-Raised by: Tim Millington +
-Assigned to: Keith Sabin (WM) +
- +
- +
-Issue No: 54 +
-Where detected (and date): eg. SVRSKAO041-20120124-1104.xml 08/​05/​2012 +
-Observation:​ Some &​lt;​Service&​gt;&​lt;​StandardService&​gt;&​lt;​JourneyPattern&​gt;​ elements do not contain a &​lt;​RouteRef&​gt;​ element. +
-Hypothesis:​ +
-Raised by: Tim Mllington +
-Assigned to: +
- +
-Issue No: 53 +
-Where detected (and date): ​ Spotted in several files, e.g. em_11-152_-1-y08.xml +
-Observation:​ An incorrect service description is present in some files, for example in file em_11-152_-1-y08.xml,​ the service description for the Service with ServiceCode 11-152-_-y08-1-T2 is as follows: +
- +
-&​lt;​Description&​gt;​to the normal timetables. The Bank Holiday timetables are subject to change at short notice - please check again&​lt;/​Description&​gt;​ +
- +
-Hypothesis: An incorrect service description is present in some files - perhaps an auto-import script populated the wrong data? +
-Raised by: Carl Partridge (FatAttitude) +
-Assigned to: Stuart Reynolds +
- +
- +
-Issue No: 52  RESOLVED +
-Where detected: I've seen this in a large number of files in different ares +
-For Example &​quot;​SVRWREO010A-20090126-5297.xml&​quot;​ from W.zip (02/​05/​2012) +
-Observation:​ Schema validation issue due to incorrect element ordering +
-Hypothesis: '​JourneyPatternTimingLinkGroup'​ elements are xsd:​sequence +
-Should be '​From'​ -> '​To'​ -> '​RouteLinkRef'​ -> '​Direction'​ -> '​RunTime'​ +
-I'm seeing '​Direction'​ -> '​RouteLinkRef'​ -> '​From'​ -> '​To'​ -> '​RunTime'​ +
-which occurs on all of the JourneyPatternTimingLinks +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: Peter Stoner +
-Peter Stoner comments: ​ This should be resolved now. +
- +
- +
-Issue No: 51  RESOLVED +
-Where detected: TNDS 1 May  +
-Observation: ​ 34 files in Scotland have not loaded  +
-Hypothesis: TXC files supplied are incomplete +
-Raised by: BaseMap +
-Assigned to: Stuart McNeill (Scotland) +
-Peter Stoner comments: Tracing the cause back it seems that the Trapeze TXC export stopped because of a missing or incorrect NaPTAN point for Ballycastle ferry terminal. +
-Stuart McNeill comments: These services will not be exported in future, because they relate to Non NaPTAN points, such as in Northern Ireland. +
- +
- +
-Issue No: 50 +
-Where detected: THDS data 27 April 2012 from all regions +
-Observation:​ data fails to convert +
-Hypothesis: data is not supported by current GTFS converters +
-Raised by: Colin +
-Assigned to: Peter Stoner +
-Peter Stoner comments: ​ This may work now that the xsd:​sequence has been sorted out. +
- +
- +
- +
-Issue No: 49  RESOLVED +
-Where detected: '​swe_32-26_-1-y10.xml'​ +
-Observation:​ Schema validation issue due to multiple occurrence of '​VehicleType'​ in VehicleJourney/​Operational. +
- +
-&​lt;​PrivateCode&​gt;​VJ_32-26-_-y10-1-1-T0&​lt;/​PrivateCode&​gt;​ +
-&​lt;​Operational&​gt;​ +
- &​lt;​VehicleType&​gt;​ +
- &​lt;​VehicleTypeCode&​gt;​DDLF&​lt;/​VehicleTypeCode&​gt;​ +
- &​lt;​Description&​gt;​Double Decker Low Floor &​lt;/​Description&​gt;​ +
- &​lt;/​VehicleType&​gt;​ +
- &​lt;​VehicleType&​gt;​ +
- &​lt;​VehicleTypeCode&​gt;​LD&​lt;/​VehicleTypeCode&​gt;​ +
- &​lt;​Description&​gt;​Double Decker Low Floor &​lt;/​Description&​gt;​ +
- &​lt;/​VehicleType&​gt;​ +
-&​lt;/​Operational&​gt;​ +
- +
-Hypothesis: The schema specifies  +
-&​lt;​xsd:​element name=&​quot;​VehicleType&​quot;​ type=&​quot;​VehicleTypeStructure&​quot;​ minOccurs=&​quot;​0&​quot;&​gt;​ +
-'​maxOccurs'​ defaults to '​1'​ as it has not been overridden here +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: Andy Hole (SW) +
-Andy Hole comments: ​ There was a type mismatch between internal and external codes. ​ This will be corrected in this week's data export. +
- +
- +
- +
-Issue No: 48 +
-Where detected: '​swe_30-225_-2_y10.xml'​ +
-Observation:​ Last journey (VJ_30-225-_-y10-2-4-T0) missing DepartureTime,​ +
-having a schema-invalid footnote with just the text '​R'​ rather than separate +
-'​NodeCode'​ and '​NoteText'​ elements. +
-Hypothesis: ​ Journey repetition record +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: Andy Hole (SW) +
- +
- +
-Issue No: 47   ​RESOLVED +
-Where detected: '​swe_30-1_-1y10.xml'​ +
-Observation:​ Conflicting bank holiday flags.  +
-The same element names appear both in both the '​DaysOfOperation'​ and '​DaysOfNonOperation'​ groups of '​BankHolidayOperation'​ +
- +
-XPath Expressions:​ +
-count(//​BankHolidayOperation[[DaysOfOperation/​EasterMonday and DaysOfNonOperation/​EasterMonday]) +
-= 182 +
-count(//​BankHolidayOperation[[DaysOfOperation/​GoodFriday and DaysOfNonOperation/​GoodFriday]) +
-= 185 +
- +
-Hypothesis: Data issue +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: Andy Hole (SW) +
-Peter Stoner comments: ​ These holidays are no longer exported in the quoted file. +
-Andy Hole comments: ​ The issue is no longer present. ​ I amended the way I do Bank Holidays. +
- +
-Issue No: 46  RESOLVED +
-Where detected: '​swe_30-1_-1y10.xml'​ +
-Observation:​ Schema validation issue due to sequence ordering of BankHoliday flag elements. +
-Hypothesis: '​HolidayMondaysDaysGroup'​ elements are an '​xsd:​sequence'​ +
-The elements are all optional but where they occur they should be ordered. +
-LateSummerBankHolidaysNotScotland -> MayDay -> EasterMonday -> SpringBank -> AugustBankHolidayScotland +
-The file contains  +
-          &​lt;​DaysOfNonOperation&​gt;​ +
-            &​lt;​GoodFriday /> +
-            &​lt;​EasterMonday /> +
-            &​lt;​SpringBank /> +
-            &​lt;​LateSummerBankHolidayNotScotland /> +
-          &​lt;/​DaysOfNonOperation&​gt;​ +
- +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: Andy Hole (SW) +
-BaseMap comment: The order remains as supplied by the regional source data. +
-Peter Stoner comments: ​ These days are no longer quoted in the export file which may be an issue in itself. +
-Andy Hole comments: ​ Issue is no longer present. ​ I amended the way I do Bank Holidays. +
- +
- +
-Issue No: 45  RESOLVED +
-Where detected: several of the SW xml files, probably all of them +
-Observation:​ Schema validation issue due to sequence ordering of Operator elements +
-Hypothesis: OperatorStructure elements belong to an '​xsd:​sequence'​ +
-In the data files NationalOperatorCode appears after OperatorNameOnLicence,​  +
-it should appear first in the Operator, before OperatorCode. +
-Raised by: Edward Wohlman (Omnibus) +
-Assigned to: Peter Stoner +
-Peter Stoner comments: this is expected to be corrected in the data release on 1 May. +
- +
- +
- +
-Issue No: 44  RESOLVED +
-Where detected: National Operator Codes April 2012 +
-Observation:​ There are some codes which are not 4 characters long, and they are not rail or air mode: AUSTP, ECCHS, EXCEL, HMV, HMPTN, WPH,  +
-Hypothesis: They may have been created in error. +
-Raised by: John Prince +
-Assigned to: The regional owners of the codes concerned. +
-Peter Stoner comments: ​ The bus codes listed above have now been amended so all bus operator codes are 4 characters. +
- +
- +
-Issue No: 43 +
-Where detected: Prerelease TNDS data, NOC +
-Observation:​ There are around 450 operators referenced in TNDS that cannot be simply matched in NOC from their region and operator code, and an additional 160 that don't seem to have records in NOC at all. See http://​eclipse.ukcod.org.uk/​~louise/​tnds/​missing_2012-04-16_with_fixes.tsv for a list of all the TNDS files that contain references to operators that can't currently be simply matched, plus a suggested matched to a NOC code, where some match could be made based on the name. +
-Hypothesis: NOC is missing some regional codes, and also some records. +
-Raised by: Louise Crow / mySociety +
-Assigned to:  Traveline Regional Managers +
-PJS: West Midlands should be ok now in latest version of NOC.  Other regions to follow. +
-Peter Stoner comments: ​ SE, EM, EA issues ​now understood. ​ Working on a solution. +
- +
- +
-Issue No: 42   ​RESOLVED +
-Where detected: TNDS test data swe_48-329A-1-y10.xml +
-Observation:​ Negative run time of -PT2M +
-Hypothesis: seems to cancel a wait time of PT2M +
-Raised by: Eduard Tulp +
-Assigned to: Andy Hole +
-Andy comments: surprised ​that the -ve time got through. ​ Data will be checked. +
-Peter Stoner comments: ​ This seems to be corrected in the file quoted. +
- +
- +
-Issue No: 41  CLOSED +
-Where detected: TNDS test data SUF_7-1_-1-y08.xml +
-Observation:​ &​lt;​Description&​gt;​Nowton � Bury St. Edmunds Town Centre � Mildenhall Rd Est&​lt;/​Description&​gt;​ +
-Hypothesis: Not entirely sure whether that is formally allowed and it is not a big issue, but our sofwtare was unhappy with it. +
-Raised by: Eduard Tulp +
-Assigned to: Stuart Reynolds +
-Peter comments: This file is not in the latest data supplied. +
-Stuart comments: I think that this was caused by Suffolk copying and pasting something that Microsoft Word had auto-corrected. It has a habit of turning a perfectly respectable hyphen into a longer em dash, that then falls apart as soon  as the character set changes. I did tell them about some, and it may be  that it is fixed already. +
-Peter comments: So we will try and avoid this happening but no auto checks in place yet. +
- +
- +
-Issue No: 40  RESOLVED +
-Where detected: TNDS test data +
-Observation:​ Files are not zipped +
-Hypothesis: Is it possible to have this data provided in a zip file to improve the download time? +
-Raised by: Jeppesen, JourneyPlan,​ ITOworld +
-Assigned to: Peter Stoner +
-Peter comments: ​ The files are expected (soon) to be in separate zip folders relating to the (regional) source of the data.  A test file of London data has been zipped. +
-Peter comments: ​ The files are now zipped. +
- +
- +
-Issue No: 39  RESOLVED +
-Where detected: ​ TNDS NPTDR format sample data +
-Observation:​ Sample files in a single directory +
-Hypothesis: an index file mapping file names to region +
-names will be useful - as this allow automated referential integrity +
-checks ​with other data sources that reference regions eg. NOC +
-Raised by: Louise Crow +
-Assigned to: Peter Stoner +
-Peter comments: ​ The files are expected to be in separate zip folders relating to the (regional) source of the data. +
- +
- +
-Issue No: 38 +
-Where detected: TNDS NPTDR format sample data +
-Observation: ​ JourneyPatternTimingLink elements don't have id attributes as they do in the beta data set. +
-Hypothesis: ​ may not make any difference to downstream loading processes +
-Raised by: Louise Crow +
-Assigned to: Peter Stoner +
-Eduard Tulp adds: In both the Wales and the YK sets the mandatory Id in the JourneyPatternTimingLink elements are all missing. +
- +
- +
- +
-Issue No: 37 +
-Where detected: TNDS NPTDR format sample data +
-Observation:​ The operator code &​quot;​554TMT&​quot;​ used in em_11-112_-1-y08.xml isn't an +
-operator code listed for any region in NOC - looks like maybe it +
-should be just &​quot;​TMT&​quot;​. Also if this is the "T M Travel Ltd&​quot;​ in NOC, the +
-OperatorNameOnLicence value of &​quot;​TM Travel&​quot;​ doesn'​t exactly match the +
-VOSA_PSVLicenseName field in NOC - which has "T M Travel Ltd&​quot;​ +
-Hypothesis:  +
-Raised by: Louise Crow +
-Assigned to: Peter Stoner +
- +
- +
-Issue No: 36 +
-Where detected: ​ TNDS NPTDR format sample data +
-Observation:​ The operator code &​quot;​JT&​quot;​ used in swe_30-218_-1-y10.xml ​ and +
-swe_30-219_-1-y10.xml isn't an operator code listed for any region in +
-NOC and the name &​quot;​Jacketts Coaches&​quot;​ doesn'​t appear in NOC +
-Hypothesis: a missing entry in NOC? +
-Raised by: Louise Crow +
-Assigned to: Peter stoner +
- +
- +
-Issue No: 35 +
-Where detected: ​ Comparison of Service 79 Barnard C to Richmond supplied to TNDS by Yorkshire and NE. +
-Observation:​ Yorkshire has &​quot;​Schooldays&​quot;​ but no dates. ​ NE has &​quot;​English shooldays only&​quot;​ and &​quot;​Schooldays only&​quot;​ both with dates but they are not current. +
-Hypothesis: This should not be a problem as a user of the data should apply more recent dates from the central Services Organisation file. +
-Raised by: Peter Stoner +
-Assigned to: +
- +
- +
-Issue No: 34 +
-Where detected: Comparison of Service 79 Barnard C to Richmond supplied to TNDS by Yorkshire and NE. +
-Observation:​ Differences in Organisation codes. ​ Yorkshrie quotes &​quot;​11&​quot;,​ Durham quotes 1000, 1011, 1002 and 0000. +
-Hypothesis: These may be local organisation codes. ​ Need to check with main Serviced Organisations file. +
-Raised by: Peter Stoner +
-Assigned to: John Prince +
-John Prince: I am querying the source of the &​quot;​11&​quot;​ code from Trapeze. +
- +
- +
- +
-Issue No: 33   ​CLOSED +
-Where detected: NOC_DB120329.xls +
-Observation:​ There are two NOC codes for the Huddersfield Bus Company - HUDB and HBSY, and for East Yorkshire Motor Services (SDMS, EYMS, same VOSA license number), The A & M Group (AMMG, AAMG), Harrogate Coach Travel (HCTY, HGCT) +
-Hypothesis: One's a duplicate, or NOC codes are not supposed to be nationally unique and represent a license holder uniquely? +
-Raised by: Louise Crow / mySociety +
-Assigned to: John Prince +
-John Prince replies to say:  +
-There are public name differences in the local dataset which should perhaps be added to NOC so that the reason for their being two codes is made clear. ​  +
-Peter Stoner comments: ​ We have agreed to include an entry for each PSV number and more than one entry if the public name needs to be different in use in different places. +
-Louise Crow comments: In this case, it will be important to be able to identify in NOC cases where two entries refer to the same entity - is the reliable way to do this by looking for duplicated VOSA license numbers? +
-Peter Stoner comments: ​ Yes, comparing the VOSA licence numbers should be a reliable method of finding which operator is the same, but the public name and feedback channels may vary. +
- +
- +
-Issue No: 32 +
-Where detected: travelineFullSet14032012.zip +
-Observation:​ There are instances where journeys for what appears to be a single service are distributed across multiple files with no simple way to relate them back together - e.g. SVRLAAX0600-20110919-10555.xml and SVRGM00X600-20110919-6209.xml appear to describe outbound and inbound journeys on one route - the X60 run by Stagecoach in Lancashire - but they don't share ServiceCodes or OperatorCodes +
-Hypothesis: Not all services whose data are distributed across files share a ServiceCode or OperatorCode. +
-Raised by: Louise Crow / mySociety +
-Assigned to: Peter Stoner +
-Peter Stoner comments: ​ We are considering merging journeys into files for the moment find it helps match to the source file if we keep them as they came in.  What do you think is the best way to group journeys? ​ By operator and line eg 79, 79A, N79 in separate files, by operator and service ie 79 (includes 79, 79A and N79). Would you include the journeys of another operator eg if they provided the evening service under contract? ​ Of course this implies we have got issue 31 resolved. +
-Louise Crow comments: For my purposes, operator and line are the most useful ways to group journeys as this maps to what the public would understand as a bus route. If you can reliably include the journeys of another operator that runs the same service under contact, that would be ideal. +
- +
- +
-Issue No: 31  RESOLVED +
-Where detected: travelineFullSet14032012.zip +
-Observation:​ There are differences in the interpretation of TransXChange elements in the dataset that make it difficult to interpret consistently - e.g. in what is considered a Line - most files contain as the line name one identifier but swe_41-SO10-1-y10.xml contains the line name &​quot;​154/​54/​54A/​54B/​54C&​quot;​ and there doesn'​t appear to be specific information that would map the journey patterns included to a specific one of these lines. +
-Hypothesis:  +
-Raised by: Louise Crow / mySociety +
-Assigned to: Andy Hole - given the example quoted is in SW data. +
-Peter Stoner comments: ​ Useful tutoring from Nick Knowles suggests that there could be more than one line used in a given TXC timetable - eg there might be 21a and 21B for different branches, or 98 and N98 for a day and night bus, or 21C and 21A for a clockwise anticlockweise journeys on a circular route. ​ However these should be listed separately so that each journey gets the number shown on the front of the bus. +
-Andy Hole replies: ​ This (the example quoted) is a fairly easy one to explain as it should not have been included in the export as it is a composite route. ​ I will try and make sure that composite routes are not included in future TNDS exports. +
-Peter Stoner comments: The observed file is no longer included in the SW export. +
- +
- +
-Issue No: 30  RESOLVED +
-Where detected: travelineFullSet14032012.zip +
-Observation:​ There are services whose mode is given as &​quot;​air&​quot;​ in the dataset. ​ See http://​eclipse.ukcod.org.uk/​~louise/​tnds/​files_with_air_as_mode.txt +
-Hypothesis: These are either air services mistakenly included, or services incorrectly assigned the mode air. +
-Raised by: Louise Crow / mySociety +
-Assigned to: Stuart Reynolds - given the number of Bedfordshire,​ Suffolk, EM and EA references. +
-SR responds: &​quot;​Air&​quot;​ should probably be wheelchair accessbile bus.  MDV are being asked to fix this. +
-PS: SVRFAAHWY0-20100707-5411.xml SVRFAAHWY0-20100707-5412.xml are the Anglesey to Cardiff air service so is correct. ​ Air can be included in Traveline but we do not undertake to include all air services. +
-Stuart Reynolds responds: ​ That this problem has been resolved and the data he processes should not normally refer to &​quot;​air&​quot;​ mode.   +
- +
- +
- +
- +
-Issue No: 29 +
-Where detected: travelineFullSet14032012.zip +
-Observation:​ There are StopPoints with a StopPointRef missing from the latest NaPTAN database. There is no way to determine the location of these stops. For a full list of affected stops, see http://​www.rome2rio.com/​extras/​Traveline14032012.xlsx +
-Hypothesis: Maybe I'm using an old NaPTAN database? I downloaded from http://​www.dft.gov.uk/​public-transportdatamanagement/​DataUser_Login.aspx on 20/​03/​2012. +
-Raised by: Bernie Tschirren / rome2rio +
-Assigned to: Peter Stoner  +
-Comment from Peter Stoner: We expect at any one time for their to be a certain number of NaPTAN stops that cannot be found. ​ Some may have come through more quickly to TNDS than to the NaPTAN downloads, and a later download of NaPTAN will enable them to be matched. ​ Others may be errors that need to be corrected at the regional or operator data sources. ​ The analysis you have supplied is useful. ​ The regions already have reporting tools to help them match up the missing NaPTAN points.+
  
  
tnds.1339792885.txt.gz · Last modified: 2024/01/31 09:48 (external edit)
Back to top
Show page Old revisions Recent changes Sitemap