v5.1

5.1

Changes in detail

v5.0...0bbee51

Added

• Add the possibility to split networks that contain simplified edges (PR #278, 9798d33, 0ed437c, 67a6651, 98ef831, 7ec4d83, 637d62a, 2c70666, 1cbc6fa, 41788ff, b042c0d, 36d23d6, 402c256, 54af2b1, 894414a, 0fe32a2)
• Add functionality for commit-message content analysis, such as NLP tools including stemming, tokenization, and lemmatization, as well as a function to search for keywords in commit messages and a function to measure the length of the messages (PR #281, 5aa4e41, 99f0638, e469d3a, 7d8fd39, ef689f7, 6e64224, f544394, dd9246b)
• Add unlist.timestamps.if.possible parameter to get.edgelist.with.timestamps which allows callers to request a conversion of the timestamps from list to vector if possible, i.e., when there are no simplified edges in the network (PR #289, 39bf1dd)

Changed/Improved

• For consistency reasons: Ensure that the values of edge attributes are always lists even when they represent singular values (PR #278, 6fae184, 416c817)
• Reduce the amount of redundantly built networks by caching network data internally. This should improve the performance of building multi-networks, especially, when parts of the multi-networks have been built before (#119, PR #282, 06a814c, 4793eab, 8ba907f, 28d2290, 3608214, b30c7f2, 1fa340d, 40cd554, 8fcc744, ca348f1, 1d233af, 5dd5fc1)
• Internally cache commit-network data and bipartite-network data similarly to how we cache network data for author-, and artifact-networks (PR #282, PR #285, 3608214, aa7e3fa)
• Ensure that configured or implicitly-enforced undirectedness in partial networks is always dominant over configured or implicitly-enforced directedness. Furthermore, ensure consistency in the directedness used for edge generation and as a network attribute, especially in networks that consist of multiple partial networks such as multi-networks (PR #282, 65ead39, a776caf, 257a1c8, 41cff01)
• Remove redundant entries from the list of allowed edge attributes and instead add event.info.1 and event.info.2 (PR #282, 1b156c1)
• Allow the issue data attributes event.info.1 and event.info.2 on network edges (PR #282, 1b156c1)
• Add a network.type parameter to get.networks in which the caller can specify the types of networks to be constructed. This improves performance in cases where not all network types are needed, such as when building multi-networks (PR #285, bc2efd6, e9a0c16)
• Rename the list.attributes parameter in add.vertex.attribute and split.and.add.vertex.attribute to flatten.values with inverted semantics and introduce documentation for it to improve comprehensibility (PR #285, 7dab04a)
• Sort author data by author.name instead of author.id when reading it from a file (PR #286, 61b538b)
• Enhance codeface testing data by ensuring that commit ids are unique between proximity and feature data and by adding commit data that includes (1) different commits that touch the same file / function, (2) commits that are authored at the same time by different authors (PR #286, 7481099, 3e53285)
• Deprecate support for R version 4.0 because of breaking dependencies (PR #281, 3dc91b1)

Fixed

• Fix a bug in construct.edge.list.from.key.value.list that could cause a crash when constructing a network where different vertices have identical associated timestamps (PR #285, d694a68)
• Fix a bug in network construction that could lead to edges having an unwanted author.name.1 attribute (PR #285, 105fec1)
• Ensure that POSIXct values are correctly handled in add.vertex.attribute, i.e., that they are not converted to numeric values (PR #285, 7dab04a, 4924ac2)
• Handle empty edges when constructing commit networks using commit-interaction data (PR #285, d5e1e48)
• Correctly retain order of commit and mail data when merging it with PAStA, synchronicity, and commit message data (PR #286, 50b9b68)
• Fix get.edgelist.with.timestamps to work correctly on networks with dates in default (list) format (PR #289, 35b34bf)

v5.0

5.0

Changes in detail

v4.4...a462006

Announcement

• coronet is not compatible with igraph versions below 2.1.0 anymore. This is due to the simultaneous deprecation of subgraph.edges and the introduction of the replacement for it, subgraph_from_edges, in igraph version 2.1.0.
• The plotting module of coronet is not compatible with ggplot2 versions below 3.5.0 anymore. This is due to the simultaneous deprecation of the scale_name parameter of discrete_scale (which is used within the function plot.network of coronet) in ggplot2 version 3.5.0.

Added

• Add commit-interaction data and add functions read.commit.interactions for reading, as well as get.commit.interactions, set.commit.interactions, and utility functions for working with commit-interaction data (PR #252, d82857f, b4fd2a2, fd0aa05, bca3576, PR #263, 849123a, 3fb7437, 170bc66, f591528, 4c4b654) as well as tests for these features (PR #252, eeba7e2, 8bb39f4, 54b6f65, 7a5497a, 7b8585f, ef72540)
• Add commit-interaction networks that can be created with create.author.network or create.artifact.network if the artifact.relation or author.relation is configured to be commit.interaction (PR #252, d82857f, 329d97e) as well as tests for these features (PR #252, 07e7ed7, 7068cfa)
• Add commit network as a new type of network. It uses commits as vertices and connects them either via cochange or commit interactions. This includes adding new configuration parameters and the function add.vertex.attribute.commit.network for adding vertex attributes to a commit network (PR #263, ab73271, ab73271, cd9a930)
• Add the possibility to split data time-based by multiple data sources (PR #261, 1088395, e1f79fc, 0bb187f, 371a97a)
• Add remove.duplicate.edges function that takes a network as input and conflates identical edges (PR #268, d9a4be4, 0c2f47c, c6e90dd)
• Add cumulative as an argument to construct.ranges which enables the creation of cumulative ranges from given revisions (PR #268, a135f6b, 8ec207f)
• Add function get.last.activity.data to compute developers' last activities in a project, as well as function add.vertex.attribute.author.last.activity to add a developer's date of last activity as vertex attribute to a network, as well as helper functions get.aggregated.activity.data and add.vertex.attribute.author.aggregated.activity to allow for other activity aggregations than first and last activity (PR #275, 9f23161, 8660ed7)
• Add four new metrics that can be used for the classification of authors into core and peripheral: betweenness, closeness, pagerank, and eccentricity (PR #276, e27acb5, 2178808)
• Add helper function for prefixing function names with file names in util-read.R (PR #252, f8ea987)
• Add line-based code coverage reports into CI pipeline. Coverage reports are generated by coverage.R (PR #262, 10cac49, b3b9f4a, c815d18, e809352, 32d0482)
• Add tests for uncovered functionality in util-misc.R and util-networks.R (PR #264, ff30f32, af80551)

Changed/Improved

• Breaking Change: Change the default representation of edge attributes from vectors to lists. This change is necessary for the interplay of coronet networks with certain igraph functionality since igraph version 2.1.0 (PR #274, 1c35d1f, eda30b8, 0c6b2eb, 44c7b72, 7303eab, 0c27012)
• Change the default value for the issues.from.source configuration parameter. Instead of reading JIRA and GitHub issues together, which was the previous default, the new default value causes only GitHub issue data to be read. To restore the previous default behavior and read data from both issue sources, this now needs to be manually configured when needed. (PR #264, 5ff83c3, 8c8080c, 8bcbc81)
• Replace deprecated igraph functions by their preferred alternatives (PR #264, PR #268, PR #274, PR #279, 0df9d5b, 7ac840d, e3617b8, 4b0d522, f29662b)
• Remove deprecated parameter of ggplot2::discrete_scale (PR #279, 027ce79)
• Deprecate support for R version 3.6 (PR #264, c8e6f45, fb3f547)
• Explicitly add R version 4.4 to the CI test pipeline (c8e6f45)
• Refactor function construct.edge.list.from.key.value.list to be more readable (PR #263, 05c3bc0)
• Update necessary igraph version to 2.1.0 in README.md (PR #274, 6c3bcd1)
• Update necessary ggplot2 version to 3.5.0 in README.md (PR #279, 027ce79)
• Include core/peripheral classification inREADME.md (PR #276, 6101e11, c6744c0, 5fc2da5)

Fixed

• Fix the creation of edgelists for issue-based artifact-networks by correctly iterating over the issue data (PR #264, 321d850)
• Add range information to network-splits when splitting a network using split.network.time.based.by.ranges. This effect also propagates into split.networks.time.based (PR #274, 87911ad)
• Fix a bug in extract.timestamps that occurs when the first data.source contains empty data and that leads to a return value of type numeric which should be POSIXct (PR #270, 10696e4, 646c01a)
• Adjust metrics.scale.freeness and metrics.is.scale.free functions to be compatible with both older and newer igraph versions (PR #274, 4b0d522)

v4.4

4.4

Changes in detail

v4.3...24005e4

Announcement

• Due to a bug in package igraph (igraph/rigraph#1158), which is present in their versions 2.0.0 to 2.0.3, the functions metrics.scale.freeness and metrics.is.scale.free can currently not be used with these igraph versions. If you need to call any of these two functions, you either need to install igraph version 1.6.0 or wait until the bug in igraph is fixed in a future version of igraph.

Added

• Add issue-based artifact-networks, in which issues form vertices connected by edges that represent issue references. If possible, disambiguate duplicate JIRA issue references that originate from codeface-extraction (PR #244, PR #249, 98a93ee, 771bcc8, 368e792, fa3167c, 4646d581d5e1f63260692b396a8bd8f51b0da48fda, ed77bd7)
• Add a new split.data.by.bins function (not to be confused with a previously existing function that had the same name and was renamed in this context), which splits data based on given activity-based bins (PR #244, ece569c, ed5feb2)
• Add get.bin.dates.from.ranges function to convert date ranges into bins format (PR #249, a1842e9, 858b181)
• Add the possibility to simplify edges of multiple-relation networks into a single edge at all instead of a single edge per relation (PR #250, PR #255, 2105ea8, a34b5bd, 3451641, 78f4351, d310fdc, 58d77b0)
• Add network simplification to showcase file (PR #255, dc32d44)
• Add tests for network simplification (PR #255, 338b069, 8a6f47b, e01908c, 7b6848f, 666d784)
• Add an assert.sparse.matrices.equal function to compare two sparse matrices for equality for testing purposes (PR #248, 9784cdf, d9f1a8d)
• Add tests for file util-networks-misc.R (#242, PR #248, PR #258, f3202a6, 030574b, 380b022, 8b803c5, 7335c3d, 6b600df, a53fab8, faf19fc)

Changed/Improved

• Add input validation for the bins parameter in split.data.time.based and split.data.by.bins (PR #244, ed0a530, 5e5ecba)
• Test for the presence and validity of the bins attribute on network-, and data-splits (PR #249, c064aff, 93051ab)
• Simplify call chain-, and branching-routes in network-splitting functions and consequently set the bins attribute on every output network-split (while minimizing recalculations) (PR #249, #256, PR #257, a1842e9, 8695fbe)
• Rename split.data.by.bins into split.dataframe.by.bins as this it what it does (PR #244, ed5feb2)
• Throw an error in split.data.time.based.by.timestamps if no custom event timestamps are available in the ProjectData object (6305adc)
• Enhance testing data by adding add_link and referenced_by issue events, which connect issues to form edges in issue-based artifact-networks. This includes duplicate edge information in JIRA data as produced by codeface-extraction (PR #244, 9f840c0, ea4fe8d, 6eb7311)
• Add a check for empty networks in the functions metrics.scale.freeness and metrics.is.scale.free and return NA if the network is empty (29418f2)
• Enhance get.author.names.from.network and get.author.names.from.data to always have the same output format. Now it doesn't depend on the global flag anymore (PR #248, d87d325, ddbfe68)
• Change util-tensor.R to correctly use the new output format of get.author.names.from.network (PR #248, 72b663e)
• Throw an error in convert.adjacency.matrix.list.to.array if the function is called with wrong parameters (PR #248, ece2d38, 1a3e510)
• Rename compare.networks to assert.networks.equal to better match the purpose of the function (PR #248, d9f1a8d)
• Explicitly add R version 4.3 to the CI test pipeline (9f346d5)

Fixed

• Reformat event.info.1 column of issue data according to the <issue-%source-%id> format, if the content of the event.info.1 field references another issue (PR #244, 62ff9d0)
• Rename vertex attribute IssueEvent to Issue in multi-networks, to be consistent with bipartite-networks (PR #244, 26d7b7e)
• Fix an issue in activity-based splitting where elements close to the border of bins might be assigned to the wrong bin. The issue was caused by the usage of split.data.time.based inside split.data.activity.based to split data into the previously derived bins, when elements close to bin borders share the same timestamps. It is fixed by replacing split.data.time.based by split.data.by.bins (PR #244, ece569c)
• Remove the last range when using a sliding-window approach and the last range's elements are fully contained in the second last range (PR #244, 48ef4fa, 943228f)
• Fix broken error logging in metrics.smallworldness (03e0688)
• Fix get.expanded.adjacency to work if the provided author list does not contain all authors from the network and add a warning when that happens since it causes some authors from the network to be lost in the resulting matrix (PR #248, ff59017)
• Fix get.expanded.adjacency.matrices to have correct names for the columns and rows (PR #248, PR #258, e72eff8, a53fab8)
• Fix get.expanded.adjacency.cumulated so that it works if weighted parameter is set to FALSE (PR #248, 2fb9a5d)
• Fix multi-network construction to work with igraph version 2.0.1.1, which does not allow to add an empty list of vertices (PR #250, 5547896)

v4.3

4.3

Changes in detail

v4.2...a0c3a52

Added

• Add function verify.data.frame.columns to check that a dataframe includes all required columns, optionally with a specified datatype (PR #231, d1d9a03)
• Add helper function is.single.na to check whether an element is of length 1 and is NA (ddff2b8, ccfc2d1)
• Add CI support for GitHub Actions (PR #234, fa1fc4a)

Changed/Improved

• Include structural verification to almost all functions that read dataframes from files or set a dataframe (setter-functions) (PR #231, b7a9588)
• Include removal of empty and deleted users in the setters of mails, commits, issues, and authors. For commits, also the committer.name column is now checked for deleted or empty users (PR #235, 08fbd3e)
• Check for empty values (i.e., values of length < 1) when updating configuration attributes and throw an error if a value is empty (9f36c54)

Fixed

• Fix check for empty input files in utility read functions. Compared to unpresent files, empty files do not throw an error when reading them, a check for nrow(commit.data) < 1 is therefore required (PR #231, ecfa643)
• Fix various problems regarding the default classes of edge attributes and vertex attributes, and also make sure that the edge attributes for bipartite edges are chosen correctly (PR #240, 4275b93, 98a6deb, b8232c0, a953555, 820a763)
• Add argument to construct.edge.list.from.key.value.list function which differentiates if constructed edges are supposed to be artifact edges, in which case we check if the artifact attribute is present for edges and replace it by author.name. (PR #238, e2c9d6c, 7f42a91)
• Change edge construction algorithm for cochange-based artifact networks to respect the temporal order of data. This avoids duplicate edges. (PR #238, e2c9d6c)
• Clarify that edges in issue-based artifact-networks are not available yet in README.md. (PR #238, 18a54f0)
• Fix bugs related to expanded adjacency matrices and update the initiation of sparse matrices to the most recent version of package Matrix, to replace deprecated and disfunct function calls. Due to this update, package versions prior to 1.3.0 of the Matrix package cannot be used any more. If the 'install.R' detects that a version prior to 1.3.0 is installed, it now automatically tries to re-install package Matrix once (PR #241, 573fab2, 2f06252)
• Prevent R warnings 'length(x) = 2 > 1' in coercion to 'logical(1)' in if conditions for updating configuration values, in update functions of additional data sources, and in get.first.activity.data() (PR #237, PR #241, ddff2b8, e1579ca)
• Prevent R warnings In xtfrm.data.frame(x) : cannot xtfrm data frames (PR #237, c24aee7)
• Fix wrong bracket in pasted logging message (PR #241, 50c68cb)
• Replace deprecated R function calls (PR #237, ed43382)

v4.2

4.2

Changes in detail

v4.1...b7db5cd

Added

• Incorporate custom event timestamps, i.e., add a configuration entry to the project configuration that allows specifying a file from which timestamps can be read, as well as an entry that allows locking this data; add corresponding functions get.custom.event.timestamps, set.custom.event.timestamps and clear.custom.event.timestamps (PR #227, 0aa3424, 0f237d0, c180398, 54e089d, 54673f8, c5f5403)
• Add function split.data.time.based.by.timestamps to allow using custom event timestamps for splitting. Alternatively, timestamps can be specified manually (PR #227, 5b8515f, 43f23a8)
• Add the following vertex attributes for artifact vertices and corresponding helper functions (PR #229, 2072807, 51b5478, 56ed57a, 9b06036, 52d40ba, e91161c)
  • add.vertex.attribute.artifact.last.edited
  • add.vertex.attribute.mail.thread.contributer.count, get.mail.thread.contributor.count
  • add.vertex.attribute.mail.thread.message.count, get.mail.thread.message.count
  • add.vertex.attribute.mail.thread.start.date, get.mail.thread.start.date
  • add.vertex.attribute.mail.thread.end.date, get.mail.thread.end.date
  • add.vertex.attribute.mail.thread.originating.mailing.list, get.mail.thread.originating.mailing.list
  • add.vertex.attribute.issue.contributor.count, get.issue.contributor.count
  • add.vertex.attribute.issue.event.count, get.issue.event.count
  • add.vertex.attribute.issue.comment.event.count, get.issue.comment.count
  • add.vertex.attribute.issue.opened.date, get.issue.opened.date
  • add.vertex.attribute.issue.closed.date, get.issue.closed.date
  • add.vertex.attribute.issue.last.activity.date, get.issue.last.activity.date
  • add.vertex.attribute.issue.title, get.issue.title
  • add.vertex.attribute.pr.open.merged.or.closed, get.pr.open.merged.or.closed
  • add.vertex.attribute.issue.is.pull.request, get.issue.is.pull.request

Changed/Improved

• Breaking Change: Rename existing vertex attributes for author vertices to be distinguishable from attributes for artifact vertices. With this change, the first word after add.vertex.attribute. now signifies the type of vertex the attribute applies to (PR #229, 75e8514)
  • add.vertex.attribute.commit.count.author -> add.vertex.attribute.author.commit.count
  • add.vertex.attribute.commit.count.author.not.committer -> add.vertex.attribute.author.commit.count.not.committer
  • add.vertex.attribute.commit.count.committer -> add.vertex.attribute.author.commit.count.committer
  • add.vertex.attribute.commit.count.committer.not.author -> add.vertex.attribute.author.commit.count.committer.not.author
  • add.vertex.attribute.commit.count.committer.and.author -> add.vertex.attribute.author.commit.count.committer.and.author
  • add.vertex.attribute.commit.count.committer.or.author -> add.vertex.attribute.author.commit.count.committer.or.author
  • add.vertex.attribute.artifact.count -> add.vertex.attribute.author.artifact.count
  • add.vertex.attribute.mail.count -> add.vertex.attribute.author.mail.count
  • add.vertex.attribute.mail.thread.count -> add.vertex.attribute.author.mail.thread.count
  • add.vertex.attribute.issue.count -> add.vertex.attribute.author.issue.count
  • add.vertex.attribute.issues.commented.count -> add.vertex.attribute.author.issues.commented.count
  • add.vertex.attribute.issue.creation.count -> add.vertex.attribute.author.issue.creation.count
  • add.vertex.attribute.issue.comment.count -> add.vertex.attribute.author.issue.comment.count
  • add.vertex.attribute.first.activity -> add.vertex.attribute.author.first.activity
  • add.vertex.attribute.active.ranges -> add.vertex.attribute.author.active.ranges
• Add parameter use.unfiltered.data to add.vertex.attribute.issue.*. This allows selecting whether the filtered or unfiltered issue data is used for calculating the attribute (PR #229, b77601d, 922258c)
• Improve handling of issue type in vertex attribute name for add.vertex.attribute.issue.*. The default attribute name still adjusts to the issue type, but this no longer happens if the same name is specified manually (PR #229, fe5dc61)

v4.1

4.1

Changes in detail

v4.0...59f4f3e

Added

• Incorporate gender data, i.e., add a configuration entry to the project configuration, add function read.gender for reading gender data, add functions get.gender and set.gender and corresponding utility functions to automatically merge gender data to the author data (PR #216, 8868ff4, bfbe4de, 0a23862, a7744b5, 6a50fd1, 413e24c, 39db315, 1e4026d)

Changed/Improved

• Add mode parameter to metrics.vertex.degrees to allow choosing between indegree, outdegree, and total (#219, ae14eb4)
• Adjust .drone.yml CI config to prevent pipeline fails: R version 3.3 is not tested any more as some packages are not available any more for this R version (ca6b474). Also another docker container in the CI pipeline is used as there are problems with the previously used docker instance (937f797)

Fixed

• Fix values in test for the eigenvector centrality as igraph has changed the calculation of this with version 1.2.7. Also put a warning that we recommend version 1.3.0 in install.R and document it in the README.md (25fb862, 1bcbca9)
• Fix the filtering of the deleted user in util-read.R to always be lowercase as the deleted user can appear with different spellings (#214, 1b4072c)
• Add check to get.first.activity.data to look for missing activity types. If no activities are in the RangeData, the function will print a warning and return an empty list (PR #220, #217, 5707517, 42a4bef, d6424c0, ca8a1b4, f6553c6)

v4.0

4.0

Changes in detail

v3.7...a656026

Announcement

• coronet now has a logo and a website: https://se-sic.github.io/coronet (#167, PR #196)

Added

• Add functionality to read and process commit messages in order to merge them to the commit data (see issue #180). Three values are available for the new attribute commit.messages in ProjectConf: none, title and messages (PR #193, 85b1d05, fdc414a, 43e1894)
• Add functions cleanup.commit.message.data and cleanup.synchronicity.data to remove commit hashes that are not any more present in the commit data from the commit message data or synchronicity data (PR #193, 98e83b0)
• Add function metrics.is.smallworld to the metrics module in order to unify checks for smallworldness (similar to scalefreeness) (PR #195, ce1f812)
• Add function metrics.vertex.centralities to metrics module in order to simplify getting a data frame containing author names and their respective centrality values (d3cd528, e7182e7)
• Add function get.data.sources.from.relations to util-networks.R which extracts the data sources of a network that were used when building it (PR #195, d1e4413)
• Add tests for the get.data.sources.from.relations function (PR #195, add0c74)
• Add logo directory containing several logo variants (PR #196, 82f9971, dc4659e, fdc5e67, 752a9b3)
• Add function preprocess.issue.data, which implements common issue data filtering operations. (fcf5cee, a566cae, 5ba6feb)
• Add function get.issues.uncached, which gets the issues filtered without poisoning or using the cache. (eb919fa)
• Add function get.issues.unfiltered to get the unfiltered issues so that these methods follow the naming scheme known from the respective methods for commits (b9dd94c, e05f344)
• Add per-author vertex attributes regarding counting of issues, issue-creations, issue-comments, mails, mail-threads, ... (like mail thread count, issue creation count) (PR #194, issue #188, 9f9150a, 7260d62, 139f70b, eb4f649, 627873c, 1e1406f, 98e11ab, a566cae)
• Add functionality that allows to read any data source at any point in time, even after splitting. In this case, the read data is automatically cut to the corresponding range on the RangeData object (PR #201, 7f9394f). Additionally, when changing the configuration parameters concerning additional data sources, the environment of a ProjectData object is no longer reset (PR #201, eed45ac)
• Add new configuration parameters commits.locked, mails.locked and issues.locked to ProjectConf which, when set to TRUE, prevent the respective getters from triggering the read of the data if it is not present yet (PR #201, 3821677)
• Add support for classifying developers on the basis of more count-based classification metrics, including mail-count, mail-thread count, issue-count, issue-comment count, issue-commented-in count, and issue-created count (issue #70, PR #209, d7b2455, 6f737c8)
• Add bot filtering mechanism, which allows removing issues/mails/commits made by bots (838855f, dcce82d)
• Ignore the "deleted user", as well as the author having an empty name "" (1a08140, 24c222a)

Changed/Improved

• Breaking Change: Rename getters for main data sources: Unfiltered date is now acquired using get.<datasource>.unfiltered, filtered data is acquired using get.<datasource> (edf19cf, e05f344)
• Add check for empty network in metrics.hub.degree function. In the case of an empty network, a warning is being printed and NA is returned (PR #195, 4b164be)
• Adjust the function ProjectData$get.artifacts: Rename its single parameter to data.sources and change the function so that it can extract the artifacts for multiple data sources at once. The default is still that only artifacts from the commit data are extracted. (PR #195, cf795f2, 70c05ec, 5a46ff4, fd767bb)
• Change the internal representation of empty data from NULL to empty data frames and adapt function get.cached.data.sources() of ProjectData which returns a vector of all data sources that are cached (including additional and filtered data sources) (PR #201, aec898e, e55d088, 24c222a); additionally, introduce new function is.data.source.cached() in util-data.R that returns a logical vector indicating which of the given data sources are cached (PR #201, b49cc5d, 491e70c, 24c222a)
• Change the threshold calculation for the classification of developers to use a quantile approach when classifying on the basis of network centrality metrics (issue #205, PR #209, PR #210, 5128252, 0d6a3a1)
• Update documentation in util-network-metrics.R and util.conf.R (PR #195, f929248, de9988c, PR #199, 059b286)
• Splitting no longer loads all (additional) data sources, but only the ones that have already been cached in the ProjectData (PR #201, 52a3014, aec898e, de1bbfe)
• Improve the documentation in util-core-peripheral.R by adding roxygen skeleton documentation to undocumented functions (issue #70, PR #209, a3d5ca7, 6f737c8)
• Change the $ notation to the bracket notation in util-core-peripehral.R (issue #70, PR #209, 6f737c8)
• Add .drone.yml to enable running our CI pipelines on drone.io (PR #191, 1c5804b)
• Not only run test suite in our CI pipeline, but also run the showcase file in our CI pipeline using test data (719a4f0, 3eb31d8)
• Add R version 4.1 to test suite and adjust missing time-zone attributes on NA vectors or empty POSIXct vectors which are correctly added as of R version 4.1 (PR #203, 6b7fb36, 98c5671, 09d11ab)

Fixed

• Fix fencing issue timing data so that issue events "happen" after the issue was created. Since only commit_added events are affected, that only happens for these. (issue #185, 627873c, 6ff585d)
• Fix the function reset.environment() of both the ProjectData and NetworkBuilder class; they now reset all the data (PR #199, de091a5, fc4c086)
• Adjust the functions update.commit.message.data(), update.pasta.data(), and update.synchronicity.data(): no warning is being printed anymore when being called by the corresponding cleanup function (PR #199, e5c60a5)
• Fix issue where the data path on RangeData objects was wrong in special cases. Introduce the (private) flag built.from.range.data.read that is set according to how the object has been created (splitting manually or reading codeface ranges) and calculating the data path accordingly (PR #199, cce9527, 917bf64, 169c034). Also add tests for this new behaviour (PR #199, ef5bac6, 3aa8e7d, d454e5a, 66ad127)
• Make splitting no longer modify the original ProjectConf, instead create a copy (e82d056)
• Fix and update outdated examples in the showcase file (473c094, 287fbfa, 0a5cce4, PR #207)
• Fix generation of Codeface range directory names from commit hashes (5c90d1c)
• Fix plotting an empty network via plot.network (03f986d)
• Fix behavior of construct.ranges when only one range has to bee constructed and sliding.window = TRUE (000314b)
• Add package reshape2 to the install script as this package is used in module `util-plot-...

v3.7

3.7

Changes in detail

v3.6...af4eaa6

Added

• Add a new file util-tensor.R containing the class FourthOrderTensor to create (author x relation x author x relation) tensors from a list of networks (with each network having a different relation) and its corresponding utility function get.author.networks.for.multiple.relations (PR #173, c136b1f, e4ee0dc, 051a5f0)
• Add function calculate.EDCPTD.centrality for calculating the EDCPTD centrality for a fourth-order tensor in the above described form (c136b1f, e4ee0dc, 051a5f0)
• Add new file util-networks-misc.R which contains miscellaneous functions for processing network data and creating and converting various kinds of adjacency matrices: get.author.names.from.networks, get.author.names.from.data, get.expanded.adjacency, get.expanded.adjacency.matrices, get.expanded.adjacency.matrices.cumulated, convert.adjacency.matrix.list.to.array (051a5f0)
• Add tests for sliding-window functionality and make parameterized tests possible (a3ad0a8, 2ed84ac, PR #184)
• Add function cleanup.pasta.data to remove wrong commit hashes and message ids from the PaStA data (1797e03, PR #189)

Changed/Improved

• Adjust the function get.authors.by.data.source: Rename its single parameter to data.sources and change the function so that it can extract the authors for multiple data sources at once. The default value of the parameter is a vector containing all the available data sources (commits, mails, issues) (051a5f0)
• Adjust recommended R version to 3.6.3 in README (92be262)
• Add R version 4.0 to test suite and adjust package installation in install.R to improve compatibility with Travis CI (40aa0d8, 1ba0367, #161)

Fixed

• Fix sliding-window creation in various splitting functions (split.network.time.based, split.networks.time.based, split.data.time.based, split.data.activity.based, split.network.activity.based) and also fix the computation of overlapping ranges in the function construct.overlapping.ranges to make sure that the last and the second-last range do not cover the same range) (1abc1b8, c34c42a, 097cebc, 9a1b651, 0fc179e, cad28bf, 7602af2, PR #184)
• Fix off-by-1 error in the function get.data.cut.to.same.date (f0744c0)
• Fix missing or wrongly set layout when plotting networks (#186, 720cc7b, 877931b)
• Fix reading of the PaStA data since the file format has changed (712bbaf, PR #189)
• Fix bug that duplicates revision set ids in the mail and commit data when merging the PaStA data and also copy-paste error when merging PaStA data to commit data (1797e03, PR #189)
• Fix bug that results in an error when there is a variable called 'c' in the R environment (de42eb2, PR #189)
• Fix bug that when applying filter.patchstack.mails() to an environment with no mail data, the mail data gets set to NULL (8261475, PR #189)

v3.6

3.6

Changes in detail

v3.5...75ae4a5

Added

• Add a parameter editor.definition to the function add.vertex.attribute.artifact.editor.count which can be used to define, if author or committer or both count as editors when computing the attribute values. (#92, ff1e147)
• Add the possibility to filter out patchstack mails from the mails of the ProjectData. The option can be toggled using the newly added configuration option mails.filter.patchstack.mails. (1608e28, a932c8c)
• Add a new file util-plot-evaluation.R containing functions to plot commit edit types per author and project. (PR #171, d4af515, aa542a2. 0a0a590)

Changed/Improved

• Add R version 3.6 to test suite (8b2a52d, #161)
• Update .travis.yml to improve compatibility with Travis CI (41ce589)

Fixed

• Ensure sorting of commit-count and LOC-count data.frames to fix tests with R 3.3 (33d63fd)

v3.5

3.5

Announcement

• Rename project to coronet (#10, 929f8ce, ac1ce80)
  • Be sure to update Git remotes and submodules to the new URL!

Changes in detail

v3.4...0e8f458

Added

• Add the constants UNTRACKED.FILE, UNTRACKED.FILE.EMPTY.ARTIFACT, and UNTRACKED.FILE.EMPTY.ARTIFACT.TYPE: Commits that do not change any artifact are considered to be carried out on a meta-file called <untracked.file>. The constant UNTRACKED.FILE is added to hold the string constant. Analogously, the constants UNTRACKED.FILE.EMPTY.ARTIFACT (currently, "") and UNTRACKED.FILE.EMPTY.ARTIFACT.TYPE (currently, "") hold the constants for any artifacts and their corresponding types, respectively, "changed" in untracked files. (11428d9, 5ea65b9, dde0dd7, 2284bbe)
• Add the public method ProjectData$get.commits.filtered.uncached: The method allows for external filtering of the commits by specifying if untracked files and/or the base artifact should be filtered (this method does not take advantage of caching, whereas the method ProjectData$get.commits.filtered does) (11428d9)
• Add the parameters commits.filter.base.artifact and commits.filter.untracked.files to the ProjectConf: In addition to the ProjectConf parameter commits.filter.base.artifact (previously called artifact.filter.base), which configured whether the base artifact should be included in the get.commits.filtered method, there is now a similar parameter called commits.filter.untracked.files doing the same thing for untracked files (11428d9, 466d8eb)
• Add parameter edges.for.base.artifacts to NetworkConf : In author networks, edges do not get constructed anymore between authors for solely modifying untracked files. For authors involved in changing the base artifact, it can be configured whether edges should be created or not using the new NetworkConf parameter edges.for.base.artifacts (c60c2f6, 466d8eb)
• Add method ProjectData$get.authors.by.data.source to retrieve authors by given data-source name (#149, 6580427, 137d833)
• Add helper function create.empty.data.frame: The function returns empty data.frames (0 rows) with correct columns and, if specified, all the correct data types. In the future, functions, that return data in data.frames, should always return data.frames of the same shape (regarding columns and data types) – especially when they are empty – because this makes later case distinctions easier or unnecessary (67a4fbe, 3513647)
• For the most common types of data.frames (data.frames of commits, mails, issues, and authors) four more utility methods are added, namely create.empty.authors.list, create.empty.commits.list, create.empty.issues.list, create.empty.mails.list, create.empty.synchronicity.list, create.empty.pasta.list as well as corresponding constants holding columns and associated data types for all these empty data.frames (5f0f529, 523daef, f8e021d, 3513647, 2f4e6f0, cd3e34a)
• Add mandatory attributes in create.empty.network if wanted (cae9d4b, cc8bd86)
• Add function create.empty.vertex.list (c00101d)
• Add tests for construction of networks without data (a4b3524)
• Add tests for construction of networks without vertices (6eb214c)
• Add a note on mailing-list threads to README (c6dca27)
• Add cutting functionality to README descriptions (fb40c50)
• Add the parameter restrict.classification.to.authors to the functions get.author.class.by.type, get.author.class.overview, get.author.class.network.degree, get.author.class.network.eigen, get.author.class.network.hierarchy, get.author.class.commit.count and get.author.class.loc.count. The parameter allows to perform classifications on a limited group of authors whose names are specified in this parameter. (2492dd0, #148)
• Add test cases for util-core-peripheral.R by adding the new file test-core-peripheral.R along with test cases (2627d6c)
• Add project-configuration parameter issues.from.source to choose if only issues from JIRA, only issues from GitHub, or all issues shall be read in (PR #159, d677949, a3e7132, ea26181). Therefore two test cases, one that reads in only JIRA issues and one that reads in only GitHub issues, are added to the issue read test (65b1acd, 2d897cb)
• Add class documentation (#157, 6e33d0a, 250f9e0)

Changed/Improved

• Always add mandatory vertex and edge attributes (#154, 0526755)
• Heavily improve addition of PaStA data (cd3e34a)
• The method read.issues in util-read.R now supports the new issue data format (PR #147, 77c750c, e04ce30, 67b818a, 4020487, 3513647). Therefore, the test issue data and all related tests are updated (39971ee, 0ec6c6c, 6a9f4ad, fda000f, 3513647)
• Rename ProjectConf parameter artifact.filter.base to commits.filter.base.artifact (PR #149, 466d8eb)
• The constant BASE.ARTIFACTS is extended by adding untracked files (i.e. the new meta-file UNTRACKED.FILE), which is now considered to be a new base artifact in the case of file-level analyses. This implies, that, in case of file-level analyses, the base artifact and the untracked files fall together, while in feature-level and function-level analyses they are treated differently (d11d0fb)
• Filtering by artifact kind (e.g. filtering out either "Feature" or "FeatureExpression") is now being done in the method ProjectData$get.commits instead of the method ProjectData$get.commits.filtered (894c9a5)
• Remove get.commits.filtered.empty and corresponding filter.commits.empty method, the functionality is now included into the methods get.commits.filtered and filter.commits respectively (11428d9)
• The private method ProjectData$filter.commits now takes parameters which configure whether untracked files and/or the base artifact are to be filtered (11428d9)
• Remove get.commits.raw, set.commits.raw and read.commits.raw functions (64a9486, c26e582)
• Add commits on untracked files to test suite (#153, d9f527c)
• In the class Conf (and its sub-classes NetworkConf and ProjectConf), default parameters are not validated anymore to avoid confusion by logging output (ec8c6dd)
• In the class Conf (and its sub-classes NetworkConf and ProjectConf), stop is called on errors during parameter updates now (ec8c6dd)
• Change shape of Vertices in the legend of plots to avoid confusion (f4fb480)
• Refactor ProjectData$get.cached.data.sources to be more concise (a4e7a21)
• Update contribution guide regarding roxygen2 conventions (#157, fbc2d54, 783ee58, 6e33d0a)
• Update README regarding mandatory edge attributes (641624b)
• Rename misleading parameter names for functions get.author.class.by.type, get.author.class.overview, get.author.class.network.degree, get.author.class.network.eigen, get.author.class.network.hierarchy, get.author.class.commit.count and get.author.class.loc.count. Most importantly, the parameter range.data was renamed to proj.data for these functions. (587ef99, 81568b1, #70)
• Remove the unused functions get.commit.count.threshold and get.loc.count.threshold. (2534d73, #70)
• The function verify.argument.for.parameter was adjusted to be suitable in more general use-cases (557bdcd)
• Do not redundantly initialize data sources when splitting (35698a1)
• Read PaStA and synchronicity data only if enabled (79bf3ca)
• Add and enforce coding convention to use 'vertices' and not 'nodes'. Most importantly, the function metrics.node.degrees is renamed to metrics.vertex.degrees. (d35ce61...