odk-x/tool-suite-x

ODK-X Tool Suite Project roadmaps, issue queue, release notes and wiki.

tool-suite-x's Issues

Incorrect template name in scanned forms

When multiple templates are selected in preferences and a form is scanned, the record in "View scanned forms" might have an incorrect template name. Template name is always set to the first template in the set of templates retrieved from PreferenceManager.

Tables--importing from config error when file doesn't exist

Originally reported on Google Code with ID 859

What steps will reproduce the problem?
1. specify a config.properties file pointing to a nonexistent csv.
2. open the app to get it to import
3.

What is the expected output? What do you see instead?
I would expect a "file not found: filename.csv" error or something along those lines.
Instead it crashes the app.

Please use labels and text to provide additional information.

Reported by sudar.sam on 2013-06-27 03:29:46

Suitcase errors on reset app server with empty server

Software and hardware versions

Suitcase ODK 2.0.0 rev 214

Problem description

Suitcase shows an error if you try to reset an app server that is empty. It doesn't crash, it just shows an error message. Additionally it prints a log message:

deleteTableDefinition: client response is 404:
java.lang.IllegalStateException: Unexpected status code: 404
	at org.opendatakit.suitcase.net.ResetTask.doInBackground(ResetTask.java:62)
	at org.opendatakit.suitcase.net.ResetTask.doInBackground(ResetTask.java:15)
	at javax.swing.SwingWorker$1.call(SwingWorker.java:295)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at javax.swing.SwingWorker.run(SwingWorker.java:334)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)

Steps to reproduce the problem

Open suitcase. Select Reset from the Upload tab. Wait for it to complete. Select Reset again. When it gets to step 2 that deletes table definiitions, you will immediately get an error.

Expected behavior

The end goal of a reset is to have an empty server, so we should show no error in this case. We should just silently continue as long as we end up with that empty server. The error implies that the server failed to be cleaned.

Tables--provide option to restore a deleted row (not during sync).

Originally reported on Google Code with ID 898

If a user is aware a row is deleted (the display of which is another issue), it would
be expected to be able to restore that row, presumably moving that row's sync state
from DELETING to UPDATING. 

Reported by sudar.sam on 2013-08-28 18:15:42

Changes to the column width will be reset when you sync

Tables 210RC1

Importing identical row ids seems to replace and erase sync state

Originally reported on Google Code with ID 925

What steps will reproduce the problem?
1. import a csv with row ids defined
2. sync
3. import the same csv appending to table

What is the expected output? What do you see instead?
I'd expect either an "already exists" exception or to ignore it.

Please use labels and text to provide additional information.

Reported by sudar.sam on 2013-10-05 22:15:55

Can break tables by calling odkCommon.closeWindow while an odkData callback is running

Software and hardware versions

Tables 210RC1

Tables--export to csv and import all columns.

Originally reported on Google Code with ID 899

Do we currently support exporting even sync_state and conflict_type columns, thus permitting
complete preservation of data across devices? This should probably be accomplished,
so that if you export a table you will lose nothing.

Reported by sudar.sam on 2013-08-28 18:16:39

Join tables aren't working in spreadsheet view

Software and hardware versions

Tables

Problem description

There's a bunch of code that detects if the table has a join and adds an "open join table" button to the mDataCellMenued dropdown, but if you click that button it's most likely broken, it doesn't use the right intent to launch the join table

ODKTables--Move all Shared Preferences into Key Value Store

Originally reported on Google Code with ID 915

All values stored in the Shared Preferences should be moved into the Key Value Store
to enable more complete app configuration transfer between phone. 

Known things that currently reside in SharedPreferences and should be moved:

1) aggregate url
2) account name used for aggregate
3) configuration timestamp
4) spreadsheet font size

There could be more, but those are the ones I can think of.

Reported by sudar.sam on 2013-09-13 21:33:43

Erroneous entries in View Scanned Forms in Scan

Steps to reproduce

1. Press Scan New Form
2. Press back

Expected behavior
No new entry added to View Scanned Forms.

Actual behavior
Entry with name of template and date of "Web Dec 31 16:00:00 PST 1969" is added to View Scanned Forms.

Environment
Nexus 7 with Android 5.0.2
Scan development trunk commit 23c5de799627dcbbef6e51fb29c4fed753d966d5

Survey user interface (cont)

When selecting a drop down menu, and then following that up immediately by selecting a multiple choice list, the multiple choice selection is displayed as selected, but when navigating to the next screen, I am stopped because the multiple choice selection is required and it is interpreted as not filled out. It seems like the first action of selecting the drop down is blocking the second action from completing.

ODK Application Designer 2.0: Undefined value not recognised

If a value is undefined (not entered yet appears to be undefined) it crashes the app designer. When I tried to use an if statement to check if the value has been entered ( x === undefined, the javascript norm) it also crashes the program since x is still undefined yet I should be able to use it in a calculation like that.

Export for tables without a Survey form

For tables without a Survey form, for example "Tea houses", when attempting to export such form. Tables used to show a toast describing the problem, now Survey is brought up and Survey shows a error dialog. Is this behavior expected?

Select option not working when keyboard is open

If the keyboard is open when you attempt to use a dropdown select one menu, the option you choose does not get processed. Evident in the first question of 'Select Examples' in Survey.

Review usage of ODKFileUtils getTablesFolder

Problem description

The getTablesFolder function creates the necessary directories and returns the path. The usage of this function should be reviewed to make sure that the directory should be created.

ODK Survey ui missing show sent and unsent forms menu item; show/hide logic needs updating

Originally reported on Google Code with ID 757

On the Send Finalized Form screen:

(1) the menu item for showing all sent and unsent forms is missing.
(2) the show/hide logic needs to be updated. 

When we have multiple forms working in a single row (in some sort of state machine
model of form edits), we need a way to identify the submissions that are eligible for
sending to the server.

It will be tricky to figure out the logic for edge cases, like when a submission transits
through 2 or more 'send' states -- do we want to recognize this and allow the publishing
of the record in the earlier streams, or just support publishing in the last appropriate
data stream.

Reported by mitchellsundt on 2013-01-31 18:42:58

ODK Tables do we need form version

Originally reported on Google Code with ID 864

Currently we're allowing forms to be specified using formId, rootElement, and formVersion.
Do we need the version? If included would we even point to the right form?

Reported by sudar.sam on 2013-07-03 00:32:18

choices sheet should only be defined on formId === tableId forms

Within the 2.0 tools suite, the choices sheet in the XLSX file should really only exist on the tableId === formId form definition. It should then be referenced in all other XLSX form definitions.

The reason for this is that the choices lists in the tableId form are the ones that get written to the properties.csv and that are available in the key-value store and via the odkData result object.

The ones defined in the formId !== tableId forms are only available via the formDef.json of those forms.

Within the 2.0 survey framework, the javascript now preferentially uses the choice lists from the tableId form (those in the properties.csv). It logs a warning if a choice list from the formDef.json is used.

ODK 2.0 sync can get confounded by filenames containing #'s

Migrated to getodk/aggregate#81 by spacetelescope/github-issues-import

The filename is used verbatim in the URL (with slashes recognized and preserved in the REST document structure).

Unfortunately, if you have emacs open on a file while you adb-push to the device, the edit buffer will be named something like #filename.ext# for the filename.ext file you are editing.

When you then reset app server and push everything up to the server, it gets stuck trying to process that name.

odkTables.addRowWithSurveyDefault and odkTables.editRowWithSurveyDefault both make survey show an error and return instead of adding/editing a row

Software and hardware versions

Tables and Survey, development branch

Problem description

calling odkTables.addRowWithSurveyDefault(null, "imnci") makes survey say "getFormFolter: formId is null or the empty string!" instead of automatically selecting the default form for the table. null in this case is the doAction struct and not the form id, that's not what's causing it

Steps to reproduce the problem

odkTables.addRowWithSurveyDefault(null, "imnci")

Expected behavior

I start adding a new row

JS crash on bad column in defaults

Software and hardware versions

survey 210c

Problem description

If you supply defaults to addRowWithSurvey, like {"some_column": "default value"} and the column "some_column" doesn't exist, undefined is passed in to fromSerializationToElementType, which throws an exception and causes the page to never load. We should check its value around like 382 in database.js and display a warning instead.

Steps to reproduce the problem

add a row with bad columns using addRowWithSurvey

Expected behavior

displays a warning

Survey feature request: Better Sliders

Originally reported on Google Code with ID 778

Current Feature:
1.THe current sliders for range type questions default to a very basic slider.

Feature request:
Sliders should have optional labels attached to points on the range, like if the range
is 0-10, you should be able to specify labels at 0,3,5,6,and 10. Then you can give
them labels like:

0=Much Less
3=Less
5 = Neutral
6 = More
10 = Much More

This is so that users can have some markers on the range to guide their selections.

Reported by mojotexas on 2013-03-01 18:01:12

Zebra print driver force closes

Originally reported on Google Code with ID 839

"I am having a little problem with this app. Every time I start the app and hit "print"
then "click OK to install printer" the app force closes. I am currently using a HTC
one x with the latest software and trying to pair a zebra mz220"

From [email protected]

Reported by [email protected] on 2013-05-31 21:21:27

[ODK 2.0 Application designer rev126] XLSX converter does not detect undefined functions

Originally reported on Google Code with ID 1139

What tool are you using?
ODK 2.0, Application Designer revision 126

What is frustrating, troubling, annoying or lacking?
If the "calculates" spreadsheet calls a function that is not defined, the XLSX converter
still compiles normally, and the error is not detected until the survey attempts to
load the screen.

How would you like it to work?
The XLSX converter should detect if the "calculates" spreadsheet calls a function that
is not defined. (This would be a nice chance if it is possible to do without impacting
performance, although it's not essential.)

NOTE: the core team has limited ability to make software enhancements.
We rely heavily on community-contributed enhancements to the tools. If this
is something you very much need, consider hiring a developer to make the
changes with the stipulation that they work with the core team to specify
the details of the change and contribute the change back to the project
under an Apache 2 license.

Reported by [email protected] on 2015-04-30 03:27:06

Locale doesnt update if you go back to default, rather than set it explicitly

Software and hardware versions

ODK 2.0 RC1, Android 5.1, Nexus 6

Problem description

When the locale is changed in Services, it doesn't propogate to Survey if you change it from Spanish to default (English). It does work if you change it to English specifically.

Steps to reproduce the problem

Open an app in Tables. Open settings and change the locale to Spanish. Observe that Tables is in Spanish, launch a survey and notice Survey is in Spanish. Now go back to Tables and change the language back to "default". Observe Tables switches back to English. Launch a survey and notice Survey is still in Spanish. Go back to Tables and change the locale to English. Launch a Survey and notice that Survey is in English now.

Expected behavior

Survey should switch to English when we go back to default, the way Tabls does.

Other information

Found with the Cold Chain demo

Checkpoint handling

On screen rotations checkpoints are automatically discarded.

In general, checkpoints are confusing. We should talk about the overall plan for them.

Protocol error on string > 255 chars

Got an odd bug syncing to aggserver.southcentralus.cloudapp.azure.com as olive. It displayed the same thing in an alert dialog.
06-05 12:14:01.216 9353-9369/org.opendatakit.services:sync E/SyncNotification: 0 FINAL SYNC Notification -default TEXT:Protocol Error.
Please verify that the server and device are compatible.
0

Also when I dismissed the alert dialog, the "SYNC NOW" button was disabled.

06-05 12:16:50.341 9353-9421/org.opendatakit.services:sync E/HttpRestProtocolWrapper: Unexpected server response statusCode: 400
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync E/ProcessRowDataPushLocalChanges: Exception in synchronizeTable - pushing data up to server on table: registration exception: Unexpected server response status...
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: org.opendatakit.services.sync.service.exceptions.ServerDetectedVersionMismatchedClientRequestException: Unexpected server response statusCode: 400
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.logic.HttpRestProtocolWrapper.httpClientExecute(HttpRestProtocolWrapper.java:904)
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.logic.AggregateSynchronizer.pushLocalRows(AggregateSynchronizer.java:551)
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.logic.ProcessRowDataPushLocalChanges.pushLocalChanges(ProcessRowDataPushLocalChanges.java:417)
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.logic.ProcessRowDataOrchestrateChanges.synchronizeTableDataRowsAndAttachments(ProcessRowDataOrchestrateChanges.java:274)
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.logic.ProcessRowDataOrchestrateChanges.synchronizeDataRowsAndAttachments(ProcessRowDataOrchestrateChanges.java:166)
06-05 12:16:50.342 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.AppSynchronizer$SyncTask.sync(AppSynchronizer.java:302)
06-05 12:16:50.343 9353-9421/org.opendatakit.services:sync W/System.err: at org.opendatakit.services.sync.service.AppSynchronizer$SyncTask.run(AppSynchronizer.java:174)
06-05 12:16:50.343 9353-9421/org.opendatakit.services:sync W/System.err: at java.lang.Thread.run(Thread.java:818)
0

Configuration should be more intelligent on fail

Originally reported on Google Code with ID 916

Configuration from the config.properties currently does not fail gracefully in the case
of errors. The timestamp of when the last configuration was attempted (which is compared
to the timestamp on config.properties to determine if a configuration attempt is necessary)
is saved at the beginning of the attempt. This prevents cyclical force closes in the
case of an error in the properties file.

The user could then not be informed that configuration only reached a partial state.

Reported by sudar.sam on 2013-09-13 21:35:25

Scan should have an example form

Software and hardware versions

Scan rev212 RC2

Problem description

Survey and Tables have a plethora of examples to use in app-designer, but Scan has none. The old example form is no longer compatible and was created before we had a "save as incomplete" option. We can probably use the VR Maternal health form from Malawi but would want to confirm with them first.

ODK Survey -- widget suggestions

Originally reported on Google Code with ID 774

Great work done on ODK build! I have some suggestions for making this tool even more
powerful and able to manage complex surveys. I don't know if some these suggestions
are possible or already supported.
1) Option for randomizing list of answers for a question.
2) Option for randomizing group of questions.
3) Option for number of answers allowed in multiple questions.
Example:
Q1: Choose the two most important reasons from the following list?

4) Option for exclusive answers in multiple questions (answers like "Don't know" or
"None of the above" that can only be selected alone).
5) Option for terminating the interview based on the answers given in a question.
6) Use of "go to" or "jump" to another question if a rule is satisfied.
7) Option for skipping or showing only answers given in previous question (apply for
questions with the same list of answers)
Example 1:
Q1: Which countries have you ever visited?
Q2: Which of these did you like the most? (only selection between countries the respondent
visited)

Example 2:
Q1: Which is your favourite colour?
Q2: What other colours do you like? (favourite colour must be skipped in Q2)

Thank you in advance.

Reported by mitchellsundt on 2013-02-25 18:52:49

Focus snaps to top of multiscreen length pages

When a page is longer than the device screen and the user scrolls down, if they respond to a prompt the screen will snap back to the top of the page. This makes it difficult to fill out long pages, because you are constantly being snapped back to the top. The issue is exacerbated by slow performance on older devices.

Verify username screen

This screen displays "anonymous" before you choose to verify permissions - rather than the username or account you selected on the settings page. This is confusing. It should perhaps show the currently recognized user and the username/account that you are attempting to verify.

conflict resolution does not allow permissions field resolution

The loader that scans the rows that are in conflict calls:

ResolveActionList.noChangesInUserDefinedFieldValues()

If the only conflicts in a row are in the permissions columns, this will return true and cause the immediate silent resolution of these rows with take-server-values actions. This silently loses the permissions change.

Additionally, if you were to fix this, the screens for resolving conflicts ignore the permissions metadata columns. As a result, when resolving a conflict, local changes to the permissions columns cannot be applied except if you 'take all local changes'; if you pick-and-choose your changes, the local changes to the permission columns will be lost.

This should probably be changed so that for a privileged user (one with "p" permissions on this row), the OdkResolveConflictFieldLoader and UI should present the changes to the permissions columns and allow that user to elect which changes to apply (and report these permissions changes as changes via the above predicate). Alternatively, these changes could be automatically applied and not presented to the user for picking-and-choosing (in which case the logic in around the noChangesInUserDefinedFieldValues call should be revised.

In either case, there is a problem in the conflict resolution logic as-is.

This same thing impacts checkpoints, but checkpoints are currently prevented from modifying their permissions columns so as not to get into a state where the active user cannot navigate through a survey. i.e., it is a hidden gotcha within that logic, but that gotcha path is not able to be exercised.

DetailWithList view bottom pane briefly flashes top pane's HTML

Software and hardware versions

ODK 2.0.0 rev 214

Problem description

When a DetailWithList view is loaded, the top half loads and the bottom half briefly flashes with the HTML from the top half before finally loading its correct HTML after setSubListView is called.

Steps to reproduce the problem

Use DetailWithList view under normal circumstances. This happens on the first load of the TableDisplayActivity

Expected behavior

The bottom half should remain completely blank until setSubListView is called.

Leading spaces injected in select_one_with_other

Software and hardware versions

Survey 210b

Problem description

Selects demo other box for color adds a leading space where there should be none

Steps to reproduce the problem

go to selects demo, enter "teal" for color, no prompt appears. Delete the space that survey added before "teal" and it will appear

Expected behavior

no leading space

Survey feature request: add clone-existing form and generate instanceName from form data

Originally reported on Google Code with ID 775

From the ODK Survey Alpha ( [email protected] ):

2) One open a template, great to get the previous forms, could an extra button be added
which was to copy/clone a form, this could be soo useful when surveys are entering
simlar data in one site etc.
3) Can you customize the form name from the actual form ie (use an ID entered by user).

Reported by mitchellsundt on 2013-02-28 17:27:45

Survey user interface

1. After inputing text into a freeform text field, if I press the next button, it will not respond. Instead, it will pop me out of the context of the text field and grey the next button, but not actually trigger the next page.

2. When a list of suggestions appears for a freeform text input, if you try to scroll to another part of the screen it keeps immediately popping you back to the text box. It doesn't do it in a way that looks good, it ends up jumping around really quickly.

Plumb local only tables up from the AIDL into the Javascript Injected Objects

Software and hardware versions

RC2 rev212

Problem description

The local only table APIs are not plumbed up to the Javascript objects and not available to users. Right now its basically just a nice to have for syncing.

DependencyChecker shouldn't hardcode values

The Dependency checker utility currently hardcodes the dependencies as well as the applications that have those dependencies.

• We shouldn't hardcode Tables dependencies into DependencyChecker. Tables should pass an argument that indicates its dependencies.
• If possible, instead of depending on OIFileManger specifically, we should see if we can depend on something more generic.

Need to get all the button, etc. text into translations.js

A substantial amount of text is still in the template files and not available as internationalized text in the translations.js file.

We should perhaps change XLSXConverter.js to look for a translations tab in the framework spreadsheet and use that to generate this file.

Color picker in the signature input is very small in landscape and on high dpi screens

Survey 210RC1

One table fails to sync

In a Survey app containing 20 individual tables, 1 of them has stopped syncing. It definitely worked initially, as there are more than 70 submissions captured, but has since mysteriously stopped. The following log text was captured from the Google cloud console debugger (the table is called 'batch'):

For now the data is being retrieved manually by exporting the csv, but it would be nice to understand what went wrong for future deployments.

Aggregate: 1.4.12
App designer: 206

06:26:57.965GET200387 B1.4 sSync 206 (gzip)/odktables/default/tables/batch/ref/uuid:12cdd9aa-2f0d-44da-b59e-27b86a02f6cd/diff?data_etag=uuid%3Af21bc4c8-e470-4309-8295-a773785cbc2a
8.22.179.4 - - [22/Jan/2017:06:26:57 -0500] "GET /odktables/default/tables/batch/ref/uuid:12cdd9aa-2f0d-44da-b59e-27b86a02f6cd/diff?data_etag=uuid%3Af21bc4c8-e470-4309-8295-a773785cbc2a HTTP/1.1" 200 387 - "Sync 206 (gzip)" "cofnav-0002.appspot.com" ms=1389 cpu_ms=969 cpm_usd=4.325e-8 loading_request=0 instance=00c61b117c4f45a4a1b583ed98751eb0f38eb3b76b0a4fa0ff8812c45798b6a3eefa0035c5cb5e app_engine_release=1.9.48 trace_id=-
{
protoPayload: {…} 
insertId: "588497040000c6e6d7628d53" 
httpRequest: {…} 
resource: {…} 
timestamp: "2017-01-22T11:26:57.965627Z" 
severity: "WARNING" 
labels: {…} 
logName: "projects/cofnav-0002/logs/appengine.googleapis.com%2Frequest_log" 
operation: {…} 
}
06:26:58.051org.opendatakit.aggregate.odktables.impl.api.wink.GaeAwareContentEncodingRequestFilter doFilter: Gae environment -- ignoring Content-Encoding header (GaeAwareContentEncodingRequestFilter.java:73)
06:26:58.051org.opendatakit.aggregate.odktables.impl.api.wink.GaeAwareContentEncodingResponseFilter doFilter: Gae environment -- ignoring Accept-Encoding header (GaeAwareContentEncodingResponseFilter.java:72)
06:26:58.147org.opendatakit.aggregate.odktables.FileManifestManager getOdkTablesUserInfo: TablesUserPermissionsImpl: uid:blue|2016-11-19T08:26:15.083+0000 (Jdk14Logger.java:87)
06:26:58.357org.opendatakit.common.persistence.engine.gae.TaskLockImpl obtainLock: Trying to get lock : 95c8afdb-7785-4c6d-8ea0-75cf1ccddc4f batch TABLES_NON_PERMISSIONS_CHANGES (TaskLockImpl.java:254)
06:26:58.358org.opendatakit.common.persistence.engine.gae.TaskLockImpl obtainLock: No pre-existing lock (TaskLockImpl.java:256)
06:26:58.483org.opendatakit.common.persistence.engine.gae.TaskLockImpl updateValuesNpersist: Persisting lock : 95c8afdb-7785-4c6d-8ea0-75cf1ccddc4f batch TABLES_NON_PERMISSIONS_CHANGES timestamp: 1485084484358 (TaskLockImpl.java:602)
06:26:58.508org.opendatakit.common.persistence.engine.gae.TaskLockImpl lockVerification: Starting lock verification (TaskLockImpl.java:508)
06:26:58.550org.opendatakit.common.persistence.engine.gae.TaskLockImpl isExpired: LockId: 95c8afdb-7785-4c6d-8ea0-75cf1ccddc4f Time left: 65809 (TaskLockImpl.java:380)
06:26:58.557org.opendatakit.common.persistence.engine.gae.TaskLockImpl obtainLock: obtainLock SUCCESS : 95c8afdb-7785-4c6d-8ea0-75cf1ccddc4f batch TABLES_NON_PERMISSIONS_CHANGES (TaskLockImpl.java:328)
06:26:58.565org.opendatakit.common.persistence.engine.gae.StringFieldLengthMapping assertStringFieldLengths: opendatakit.__ODKTABLES__TABLE_DEFINITIONS3 (StringFieldLengthMapping.java:123)
06:26:58.598org.opendatakit.common.persistence.engine.gae.StringFieldLengthMapping assertStringFieldLengths: opendatakit.__ODKTABLES__COLUMN_DEFINITIONS3 (StringFieldLengthMapping.java:123)
06:26:58.640org.opendatakit.common.persistence.engine.gae.StringFieldLengthMapping putEntity: opendatakit.__ODKTABLES__COLUMN_DEFINITIONS3 (StringFieldLengthMapping.java:91)
06:26:58.941org.opendatakit.common.persistence.engine.gae.StringFieldLengthMapping assertStringFieldLengths: opendatakit.__ODKTABLES__TABLE_BATCH (StringFieldLengthMapping.java:123)
06:26:58.962org.opendatakit.common.persistence.engine.gae.StringFieldLengthMapping assertStringFieldLengths: opendatakit.__ODKTABLES__TABLE_BATCH_LOG (StringFieldLengthMapping.java:123)
06:26:58.979org.opendatakit.aggregate.odktables.DataManager revertPendingChanges: Reverting changes for dataETag uuid:1a58f6c6-58f3-4151-ab2a-2bb0e0aef9f4 (DataManager.java:156)
06:26:59.240org.opendatakit.aggregate.odktables.DataManager getSequenceValueForDataETag: Multiple records for dataETagAtModification uuid:f21bc4c8-e470-4309-8295-a773785cbc2a count: 30 (DataManager.java:545)
06:26:59.300org.opendatakit.common.persistence.engine.gae.TaskLockImpl deleteLock: removing lockId: 95c8afdb-7785-4c6d-8ea0-75cf1ccddc4f SUCCESS : deleteLock : batch TABLES_NON_PERMISSIONS_CHANGES (TaskLockImpl.java:226)
06:26:59.305org.opendatakit.common.persistence.engine.gae.TaskLockImpl releaseLock: releaseLock SUCCESS : 95c8afdb-7785-4c6d-8ea0-75cf1ccddc4f batch TABLES_NON_PERMISSIONS_CHANGES (TaskLockImpl.java:494)

data('nonexistent_field') should alert an error

Software and hardware versions

odk2 210c

Problem description

calling data('some_field_that_doesnt_exist') in a query filter or an assign doesn't display a warning, it should

Steps to reproduce the problem

calling data('some_field_that_doesnt_exist')

Expected behavior

alerts an error

Custom home screen switch for Tables doesn't work

Custom home screen is always enabled. Force stopping the apps and resetting configuration also don't turn off custom home screen.

ODK Survey select_one_with_other is not working

select_one_with_other prompt hangs the form when used.
The "selects" example (the color prompt) makes use of this

Plumb up order by string arrays to javascript layer

Software and hardware versions

RC2 Rev 212

Problem description

Currently the database layer supports orderbycololumn and orderbydirection as string arrays, but the javascript only accepts single string. We should plumb the string array support up to the javascript layer.

Tables--display in the UI that deleted rows need to be synched

Originally reported on Google Code with ID 897

Currently we don't indicate to a user that deleted columns exist that must be synched.
We should.

Reported by sudar.sam on 2013-08-28 18:14:39

Pulling a select multiple or select one from the database will return the raw value and there's no way to get a translation.

Software and hardware versions

Tables, odkCommon

Problem description

If I have a select one question for a health facility type, and the database value for one of the choices is national_vaccine_store, it's easy to get that exact string out of the database. However to get an english translation, like "National Vaccine Store", or a spanish translation, like "Tienda Nacional de Vacunas", you need to parse the xlsx section of the formdef, there's no way to do it via odkCommon

Steps to reproduce the problem

Expected behavior

Other information

Things you tried, stack traces, related issues...

Add proper messaging for Entity Too Large errors on sync

Software and hardware versions

Services 2.0.0 rev 214

Problem description

We currently report "Device configuration failure" when we try to sync with a file that is too big for the server to handle. Instead we should have a more precise error that one of the users files is too big. The nginx returns the correct 413 Entity Too Large error. Tomcat probably does too but we would need to verify. But it is not handled on the Services side to show the specific error.