GIPPtools — Release History

The new cubeaux utility reads secondary recording channels and writes the data as ASCII table.

Updated the leap seconds list based on Bulletin C number 68 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28, 2025.

The mseedcut utility did not always retain information about previously applied timing corrections. (Only the first and last record of the resulting miniSEED file were affected by this problem.)

The (internal) format of the system event block in Cube files has changed over the years. This bugfix ensures the correct handling of all variants of the system event block.
This problem surfaced while hunting for "ring buffer overflow" errors, which are reported as system event.

The cubeinfo utility supports and lists additional block types (option --format=SUMMARY) providing information about e.g. operating voltage and temperature. (These blocks are only generated by newer Cube hardware/firmware.)

Expanding the internal subsample timing correction lists to cover additional analog-digital-converter hardware and more sampling rates.

Updating the leap seconds list based on Bulletin C number 67 (published by the Earth Orientation Center of IERS). The new leap second list will expire December 28, 2024.

Initial support for new Cube hardware/firmware.

Updated the leap seconds list based on Bulletin C number 66 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28, 2024.

Identify spurious GPS fixes recorded by Cubes deployed underground (without signal reception).

Correctly detect all "buffer overrun" system events in Cube files and only consider samples values in the file up to the data gap.

A --dry-run command line option was added to the mseedcut utility.

Updated the leap seconds list based on Bulletin C number 65 (published by the Earth Orientation Center of IERS). The new leap second list will expire December 28, 2023.

Added the command line options --select-station and --select-channel to the mseedinfo utility.

Updated the leap seconds list based on Bulletin C number 64 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28, 2023.

None of the GIPPtools will crash anymore when encountering miniSEED records containing ASCII encoded (meta-)data.

Two new command line options (--select-station and --select-channel) were added to the mseed2ascii, mseed2pdas and mseedcut utilities. Use them to select a subset of the miniSEED input for processing when working with multiplexed miniSEED files. Records containing data from other stations or recording channels are silently skipped/ignored during processing.

Updated the leap seconds list based on Bulletin C number 63 (published by the Earth Orientation Center of IERS). The new leap second list will expire December 28, 2022.

Correcting a subsample timing inaccuracy in the postprocessing (conversion) of Cube data streams.

Correctly warn the user about a missing JRE under Windows.
Previously, Windows user were only receiving a rather misleading error message when no Java Runtime Environment (JRE) was found.

Extended the mseed2mseed utility to handle the timing quality field located in blockette #1001 (inside the miniSEED header).

Updated the leap seconds list based on Bulletin C number 62 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28, 2022.

The GIPPtools now expect at least a working Java virtual machine (also sometimes called a Java Runtime Environment or simply JRE) supporting Java 8 or newer!
This change became necessary as it becomes increasingly difficult to generate Java class/jar files on up-to-date platforms that are still backward compatible back to Java Release 5!

Improving the check for corrupted Cube files. The algorithm originally used when this check was introduced with the previous GIPPtools release produced to many false alarms when the GNSS reception is not optimal.

Support Cube logger with build-in u-Blox GNSS receiver board as time source.

Updated the leap seconds list based on Bulletin C number 61 (published by the Earth Orientation Center of IERS). The new leap second list will expire December 28^th, 2021.

Cube conversion tools will ensure that the time between consecutive Cube time tags "flows" in the right direction.
This additional check was inspired by a corrupted Cube file read from a failing drive. It somehow contained fragments of an earlier recording and converting the data to miniSEED resulted in misleading error messages and other "surprising" results.

The --include-pattern=GIPP filename filter now accepts month numbers larger than '12'!
There is a new option in the Cube firmware that forces a strict increment of the month number for each consecutive month. If used in a recording that starts in December then January would become '13', February would be '14', and so on. The option is intended for long-running (> 1 year) continuous recordings where otherwise filenames might repeat. (Cube filenames do not include a year.) Cube filenames beginning with such higher month numbers should nevertheless be included by the GIPP filename filter, hence the change.

Corrected the handling of Cube "system event" block containing more than two bytes. Previously, cubeinfo would stop working when encountering one of those rare blocks. (This bug did not affect other Cube related utilities.)

Automatically add a blockette #1001 to the miniSEED record if necessary. A blockette #1001 adds (among other things) microsecond resolution to the start time of the miniSEED record. This will improve the timing precision in situations where the miniSEED output requires more resolution after processing
An example would be a 300 Hz recoding, where some recording times contain repeating decimal numbers (e.g. 14:20:09.953333…​; with an infinite number of "threes" in the fractional second part).

Updated the leap seconds list based on Bulletin C number 60 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28^th, 2021.

The mseed2segy utility no longer crashes when the miniSEED input is fragmented and contains gaps/overlaps. This only happened under very scarce circumstances.

Updated the leap seconds list based on Bulletin C number 59 (published by the Earth Orientation Center of IERS). The new leap second list will expire December 28^th, 2020.

New --bugfix-gps-epoch command line option for the mseed2mseed utility. It can be used to fix incorrect recording times that were caused by a GPS week number rollover (WNRO) problems. Please see the program documentation for more details on how to apply the corrections.

Updated the leap seconds list based on Bulletin C number 58 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28^th, 2020.

The miniSEED utilities stumbled over partially read blockettes. The problem surfaced for the first time when the mandatory blockette #1000 is placed after an optional blockette #100 (i.e. parts of blockette #1000 extend beyond the initially read 64 byte).

Preliminary read support of blockette #100 ("sample rate"). If a miniSEED record contains a blockette #100, the sample rate information from the blockette will supersede the sample rate information from the fixed header area.

Improved I/O speed when converting Cube data to miniSEED/ASCII.

Updated the leap seconds list based on Bulletin C number 57 (published by the Earth Orientation Center of IERS). The new leap second list will expire December 28^th, 2019.

Detect corrupted Cube files containing very large blobs (i.e. >8 MB) of "zero" bytes. The GIPPtools will attempt to convert as much data as possible up to the location of the first corrupt area in the file and then skip ahead to the next input file.

Detect Cube file corruption resulting from a sample block containing a few bytes to many. (Previously, only the more common case of a few missing bytes in a Cube sample block was detected reliably.)

GIPPtools incorrectly assumed that blockette #1000 is always the first blockette following the fixed header section of a miniSEED record.

Huge time spans are now parsed correctly. This affected e.g. the --shift-time option of the mseed2mseed command when shifting the start time of a miniSEED record by more than 68.1 years. (Cause of the problem was an undetected integer overflow.)

Changed the Microsoft Windows start scripts to correctly set the ERRORLEVEL return code variable after the execution of a GIPPtools utility.

Updated the leap seconds list based on Bulletin C number 56 (published by the Earth Orientation Center of IERS). The new leap second list will expire June 28^th, 2019.

Corrected a bug in the Windows start scripts that caused the utilities not to start at all if the path to the GIPPtools installation directory contained spaces or other "problematic" characters.

Improved error message when the Cube header block contains illegal (non-ASCII) characters like German "Umlaute" or French accents. (Note: Non-ASCII characters are still illegal in the Cube header!)

Updated the leap seconds list based on Bulletin C number 55 (published by the Earth Orientation Center of IERS). The new leap second list will expire Dec 28^th, 2018.

Corrected an obscure "array index out of bounds" error in the date/time computations. The code parsing the GPS time information did not correctly handle the situation of a leap second event being recorded on Dec 31^st of any year (opposed to events occurring on Jun 30^th, which worked flawless).
This error only shows if the Cube GPS unit is actually powered on during the leap second event.

The mseed2pdas utility now correctly identifies miniSEED files recorded by an EDR-209/210 units.

Reduced the memory consumption (RAM) of the cube2ascii and cube2mseed utilities by roughly 30% at the price of a slightly increasing runtime by about 3%.

Added a new timing "quality control" algorithm (--timing-control=FAKE) to the utilities cube2ascii, cube2mseed and cube2segy that replaces all Cube timing information by a made-up "fake time".

Detection of duplicate miniSEED records when rescuing data from corrupted media with the mseedrecover utility. This is the new default behaviour but can be switched off using the --no-dedupe command line option.

New a command line option to fix certain faulty record start times encountered in miniSEED files that were recorded by an EDR-209/210 data logger running a specific, buggy firmware. Please see the mseed2mseed manual page for details.

Support the new (and still experimental) event recorder units designed to record the trigger time of seismic sources. Specifically, the new utility cubeevent can be used to read detected/recorded events from files. In addition, the other Cube related GIPPtools (cubeinfo, cube2ascii, etc.) were upgraded as well to correctly handle event recorder files. (This works because event recorder write a variant of the Cube file format.)

Added a config subdirectory to the GIPPtools distribution. In this GIPPtools release it only contains a copy of the latest leap-seconds.list file. In the future it will also contain things such as default settings or user preferences.

New --sysinfo command line argument for all GIPPtool utilities. Using it will generate a brief report of the (GIPPtools relevant) computing setup you are using. The system summary should be helpful for diagnosing configuration problems or when reporting bugs to the developer.

Added an expiration date to the internal leap second table so that GIPPtool utilities will not accidentally apply an outdated leap second correction. As a consequence, the Cube conversion tools (cube2ascii, cube2mseed and cube2segy) will refuse to process a file if the available leap second information is outdated (with respect to the Cube input). The cubeinfo program will output a warning. Utilities processing miniSEED formatted (input) data are not affected.

The mseedrecover utility will (again) read directly from a Unix/Linux device file. (There was an overeager safety check complaining about a device file of "zero length".)

Implemented a workaround for the "Could not parse TAIP message…​" error message. This bug only affected users that (still) use an outdated Java 1.6 runtime to convert files recorded by Cubes with build-in "newer" GPS board. This bug did not surface when using Java 1.7 or newer.

Added a proper error message when reading from miniSEED files that contain an uninitialized "record size" header field (i.e. claim an impossible miniSEED record size of zero).

Corrected a faulty range check when writing miniSEED files using IEEE single precision floating-point encoding.

Repaired the --record-size command line option of the cube2mseed utility, which was rejecting numbers. Cause was an internal integer overflow when comparing the given record size with a build-in, upper limit.

Updated the cube2segy utility to use the same Cube I/O subroutines as cube2ascii and cube2mseed. Due to the (internal) change the utility will be more resilient to bad GPS reception. This should also improve timing precision on subsample level.

The Cube conversion tools (cube2ascii, cube2mseed and cube2segy) contain a new (default) algorithm to better quality control the recorded timing information (option --timing-control=LLS).
The main advantage of the LLS algorithm is its flexibility. It was designed to adapt to different situations and to handle different time keeping hardware as well. However, you can use the command line option --timing-control=RULE to fall back to the previously used timing quality control method.

Added a --shift-time command line option to the mseed2mseed utility moving the start time of miniSEED records by the given time span.

Upgraded the checksum calculation (option --format=CHECKSUM) of the mseedinfo utility to use the MD5 algorithm provided by the Java runtime. If accessing the MD5 algorithm fails for some reason, it will automatically fall back to SHA-1 or the previously used CRC-64 algorithm.

Slightly modified the GPS output format (--format=GPS) of the cubeinfo utility. The reported leap second information now indicates the origin of that information. The string gps-leap stands for leap second information received from the GPS satellite broadcasts, while iers-leap indicates leap seconds as officially announced by the International Earth Rotation and Reference Systems Service (IERS). The later are kept in an internal table.

Renamed the DUMP output format of the cubeinfo utility to DEBUG, which better indicates its intended use.

Updated internal leap second table with the upcoming leap second event on 2016-12-31.

Handle the case of a misspelled input file or directory more gracefully (i.e. with a proper warning message instead of just skipping the non-existing file silently).

The mseed2mseed utility erroneously did not correctly re-encode an input file containing more than 64 miniSEED records. Everything worked fine for the first 63 records, however, the following 63 records were not re-encoded at all. (No data was lost, but the output files contained the following records in their original miniSEED encoding!) And yes, this is a follow-up to the (apparently incomplete) bug fixed in the previous release. Sorry!

The mseed2mseed utility erroneously produced data gaps in the output when an input file containing more than 64 miniSEED records.
This only happened while re-encoding the data (using at least one of the options --record-size, --encoding and --byte-order). Basically, an internal buffer was not flushed (completely) when the end of an input file was reached, resulting in missing data in the output.

Handle the case of a zero length input file more gracefully, (i.e. with a proper warning message instead of a null pointer exception).

Something went wrong while packaging the previous 2015.163 release. Unfortunately, Java classes were compiled without the correct backward compatible setting. The resulting GIPPtools release required at least a Java 1.7 runtime. A quick re-compilation with correct settings solved the problem.
Release 2015.166 is functional identical to 2015.163 with the single exception that it requires only a Java 1.5 runtime environment!

Took care of a problem, where creating files of equal length (e.g. when running mseedcut --file-length=DAY) would occasionally "forget" a single sample value in the data stream. This error occurred only when a very specific combination of sample amplitudes and miniSEED record layout was encountered, which is probably the reason it went undetected for such a long time.

Took care of another problem while creating files of equal length (e.g. when running mseedcut --file-length=DAY). Once in a while a file of twice the desired length would be generated instead of two separate files. This error happened when the first sample of a new file was also located at the end of a miniSEED record in the input. Chances for this to happen depend on the used miniSEED encoding but are typically in the range from 1:1000 to 1:2000.

Utility mseedcut now correctly handles event files (option --events=file) as exclusive source for user requested time windows. Before, it would only process an event file when an additional time window was requested via other command line parameters.

All Cube related utilities now correctly handle the error case of two time tags following directly after each other in the recorded data stream. (Expected is approximately one second of samples in between time tags!)

Due to a firmware bug in (some) Trimble GPS boards, affected Cube recorder erroneously compensate a future leap second the moment it is announced via satellite and not starting at the time when the new leap second will happen. Unfortunately, pending (new) leap seconds are usually announce via GPS satellite many weeks in advance of the actual change…
A workaround was added to the GIPPtools to compensate for the premature leap second correction.

Removed a bug, where an uninitialized internal variable accidentally would prevent the display of an error message.

A new cube2ascii utility working just like the already existing cube2mseed but outputting the Cube trace in ASCII text format instead of miniSEED format.

New IEEE double (--encoding=FLOAT-64) and single (--encoding=FLOAT-32) precision floating-point numbers to the supported miniSEED encodings!

Due to some internal changes, the checksum calculation for miniSEED files (command mseedinfo --format=CHECKSUM) had to be changed.
Do not compare checksum calculated by an earlier GIPPtools release with checksums calculated with the mseedinfo tool of this (and future) releases! All checksum must be re-calculated!

The cube2mseed utility was completely rewritten from scratch! The new version will correctly handle Cube data streams spanning over several files, long(er) periods without GPS reception and compensate for drifting clocks by resampling the Cube input.

Updated all Cube related GIPPtools so that they can handle the recent additions to the Cube data format. Especially the cubeinfo program now reports on auxiliary channels as well as event and debug blocks when called with option --format=SUMMARY or --format=DUMP.

Took care of a stupid decoding error when reading Steim compressed miniSEED files written by the new EDR-209/210 logger from EarthData.

Ignore GPS information read from Cube files when the latest corresponding GPS fix is older than 10 seconds. (Note: This patch is only relevant for Cube recorders with the new GPS board actually providing that information. Cube files written by recorders with first generation GPS boards cannot provide the necessary GPS information in the first place. No need to reprocess old files!)

Made cubeinfo more resilient to crashes when working with damaged or incomplete Cube data files.

New output format called QUALITY to mseedinfo that provides information about the data quality flag and (if available) the timing quality value stored in a miniSEED record. The same information was also added to the HEADER output format of mseed2ascii.

Updated the cubeinfo tool, so it knows about the newest enhancements to the Cube hard- and software (such as e.g. auxiliary recording channels, and the new GPS board).

New environment variable GIPPTOOLS_OPTS that can be used to fine-tune the Java runtime. This would typically be used to set the Java heap size available to GIPPtool programs.

Finally, got around to write a proper Readme file and to start a Frequently Asked Questions (FAQ) list.

Expanded the GPS output format of the cubeinfo tool (command line option --format=GPS) to give an idea about the GPS reception and the quality of the Cube time information. Additional values (elevation and temperature) become available when the used Cube recorder contains one of the new GPS boards providing that information.

Took care of a problem where mseedcut failed to produce files of equal length (option --file-length) if no sample was recorded precisely on the hour. This became an issue when working with converted files that were originally recorded by a Cube.

Mseed2mseed now understands the keyword RESET when parsing actions in a rule file or at the command line. Use the new keyword to clear/remove ID values in miniSEED header fields.

The utilities cube2segy and mseed2segy can now cache the (internal) file index that is build from the input files. Using the new --index-cache option is especially helpful If you plan to use one of the programs repeatedly on the same, large dataset.

Improved syntax checking of the project file (utilities cube2segy and mseed2segy) enables better user feedback (i.e. error messages) when the format of the project file is almost but not quiet correct. Also added a few integrity checks to catch obvious typing errors.

Added an extra test to mseedrecover to make sure all samples (and not only the header fields) of a recovered miniSEED record are readable as well. Before this bug fix, some only slightly damaged records were not detected reliably.

New --events option for the mseedcut utility. Use this option to cut out more than one time window during a single program run.

Renamed EDL/miniSEED related GIPPtools by replacing the string edl with mseed in the respective command name:

The mseedrename utility will now try to preserve the "last modified" time of a file while renaming it.

The change introduced to the cube2segy and mseed2segy utility in the last 2011.166 release was reverted after user protest. By default, both tools behave again as accustomed when writing seismic sections, creating a separate file for every shot/receiver gather requested.
Use the new command line option --force-concat when all resulting seismic sections should be written to a single output file.

Instead of automatically concatenating the results from some operation to generate as few files as possible, the new default file creation strategy for GIPPtools is to start a new output file for every read input file. Judging from user feedback, this seems to be the expected behaviour by most users.
So far, this change has been implemented for mseedcut, mseedrecover, mseed2ascii, mseed2mseed and mseed2pdas. You can use the new command line option --force-concat to return to the former, concatenating output behaviour.
A nice side effect of the re-written output routines is that the utilities need (much) less memory and are more stable when applied to very large datasets.

Renamed the --segment-length command line option of mseedcut to --file-length. The new name better describes the effect the option (writing files containing data of given length).

Renamed the --output-sort command line option of mseedrecover to --force-sort. The new name aligns better with the other output options --force-overwrite as well as --force-concat and should be easier to remember by the user.

Renamed the --rule-file command line option of mseed2mseed to --rules and renamed the --event-file command line option of cube2mseed to --events. The new names are shorter and better describe what the respective (file) arguments contain.

The utility edlcut no longer crashes when the miniSEED input does not contain any relevant data for the requested time window. (This happened only under certain rare conditions.)

New options --byte-order, --record-size and --encoding in the cube2mseed utility so that the user can influence the miniSEED output format.

Instead of creating a new output file for every shot/receiver gather requested by the user, the cube2segy and mseed2segy utility will write all seismic sections into one single file. This was changed to ease the data import into other software packages. (You only need to import one single file instead of many files.)

The cube2segy utility will no longer crash when the time window of the trace to read starts before the beginning of recording in the corresponding Cube file. (This only happened when there was no data for the beginning of the time window but data for the end of the time window was available.)

Sped up the cube2segy utility to run two to five times as fast as before! Please note: The individual speed-up highly depends on the used Cube recorder settings, especially sampling rate and GPS cycling times. So, your mileage may vary…​

Upgraded cube2segy to work with the extended, multi-file Cube data format of the next generation three channel units.

The --include-pattern option can now be used more than once in the same command line. The given include patterns are cumulative.

The cubeinfo will no longer produce "NullPointerException" errors in the default INFO mode when reading from Cube files that were cut off prematurely (e.g. running out of power while recording).

The edlrename utility will no longer overwrite already existing files. Instead, a new filename with an additional number in between basename and extension will be used for writing. To get the previous behavior of overwriting existing files use the new command line argument --force-overwrite.

New OVERVIEW output format to edl2ascii that will summarize miniSEED input streams consisting of many miniSEED records in an easy (for humans) readable format.

Improved the utilities that convert to SEG-Y (cube2segy and mseed2segy). Instead of simply beginning the traces with the first sample after the respective start time, the sample closest to the start time is used. This reduces the timing error down to at most plus/minus half a sampling period. However, it also produces traces where the first sample lies before the requested window start time!
The difference between the requested beginning of the trace and the time of the first sample is documented in the SEG-Y trace header (in field "lag A").

Improved cubeinfo to work with extended Cube file format of the next generation three channel units. Due to these changes the output format of the report generated by cubeinfo has changed.

The edl2pdas, edl2ascii, cube2mseed and cubeinfo utilities will no longer overwrite already existing files by default. Instead, a new filename with an additional number in between basename and extension will be used for writing. To get the previous behaviour of overwriting existing files use the new command line argument --force-overwrite.

Times given with more than six digits for the fractional part of the second (e.g. with nanosecond resolution) are now rounded to the closest microsecond instead of truncating the fractional part completely.

Got rid of an internal integer overflow when extracting very large time windows (e.g. one year) from a miniSEED stream. This problem affected all utilities that use time windows to select data.

New cube2segy utility working just like the already existing mseed2segy but reading Cube files as input instead of miniSEED files.

Support the newer XDR style output format of Seismic Unix for the cube2segy and mseed2segy utility. (Before, only the old "native binary" format was supported.)

Added a GPS output format to the cubeinfo utility reporting time and position of all GPS blocks contained in a Cube data stream.

Made edlrecover more robust when recovering "almost-miniSEED files" (i.e. files that closely resemble valid miniSEED but still contain only binary garbage should no longer confuse edlrecover). Changed the output subdirectory name to contain the <year> as well as the <day-of-year> when using the --output-sort option.

The edl2segy utility is now called mseed2segy.

The cube2mseed utility no longer writes miniSEED files that contain gaps or overlaps. Instead, a new file will be started if a discontinuity in the data stream is detected.

Improved tolerance of the Steim decoding subroutines. The GIPPtools utilities now can read more miniSEED variants than ever before.

The edl2segy utility produced defect SEG-Y files under Solaris when using a Sun Java 1.6 virtual machine. (Strangely, it always worked with a Java 1.5 runtime.)

Occasionally, edl2segy produced traces would contain one additional sample.

Edl2segy works again! The previous release contained a subtle error that rendered edl2segy completely unusable! Fortunately, none of the other utilities was affected.

A change in the sample rate between two consecutive miniSEED records was not always detected correctly.

The edlcut and edl2edl utilities will no longer overwrite already existing files by default. Instead, a new filename with an additional number in between basename and extension will be used for writing. To get the previous behaviour of overwriting existing files use the new command line argument --force-overwrite.

Completely rewrote the internal handling of miniSEED streams. Records are no longer decoded and encoded at every opportunity, which should speed up many GIPPtool utilities.

Support Cube data loggers. Use the new cubeinfo tool to get a textual summary about the content of a Cube file and cube2mseed to convert Cube files to miniSEED format.

Support miniSEED blockette #1001 in all GIPPtool utilities. This makes it possible to work with microsecond precision (instead of "only" with 0.0001s resolution).

New CHECKSUM output format for the edl2ascii program. Use it to calculate a checksum over the samples of a continuous trace. MiniSEED header values are not included into the checksum calculation so that different miniSEED encoding scheme should still result in the same checksum, provided that the time series are identical.

The sample rate (of a miniSEED file) was added to the list of possible template variables of the edlrename utility.

The edlcut utility can now split a long miniSEED data stream into segments of given length. This can be used to produce e.g. "day files" out of a collection of input files.

New command line parameters (--stop-time, --trace-offset) were added to the edlcut utility to simplify the specification of a time window for cutting out samples from miniSEED data streams.

Error messages do no longer begin with the text SEVERE. Instead, they start with ERROR! After all, that is what they are. (Several users assumed that SEVERE is just some sort of fancy warning, which can be ignored.)

Changed an internal filename filter that falsely discarded the dot (.) as an invalid (input) directory name. It is again possible to use the current working directory (aka .) as starting point for a recursive search for input files.

New edl2edl utility, which can be used to selectively modify miniSEED header fields such as e.g. station, channel and network ID. You can also use it to change the record size, byte order or encoding scheme of a miniSEED file.

Added Steim-2 encoding support to all GIPPtool utilities.

Added a new GMT output format to the edl2ascii program. Files created using the GMT parameter can directly be used as "multiple segment" input to the psxy command of the Generic Mapping Tools (also known as GMT, hence the format name).

Improved the HEADER output format of the edl2ascii program to also report on record size, byte order and used data encoding scheme.

New edlrename utility to systematically rename miniSEED files using a user provided "filename template".

Created a separate Microsoft Windows distribution of the GIPPtools software.
The main difference between the Windows and Unix distribution is the used start script. For Unix a Bourne shell script is used while the Windows distribution relies on a Windows command file to start the various GIPPtool utilities.

The gipptools start script (jointly used by all GIPPtool utilities) will detect a set GIPPTOOLS_JAVA environment variable and use the respective Java Runtime Environment (JRE) instead of the automatically detected one.

New edlrecover utility to recover miniSEED data from damaged files or corrupted file systems.

Added Unix manual pages for edl2ascii, edl2pdas, edlcut and edlrecover.

The edl2ascii output format SUMMARY was renamed to INDEX, which is more descriptive and frees the term SUMMARY to be used for the new output format.

Added a new SUMMARY output format to edl2ascii that will summarize miniSEED input streams consisting of many miniSEED records.

Significantly reduced the memory requirements of edl2segy. The program now runs comfortable in less than 512 MByte RAM and does not longer crash on larger jobs.

Removed a command line parsing error from the program edlcut to make it possible to cut out time series data of less than ten seconds length. Before the fix edlcut demanded erroneously at least a two-digit number for the --trace-length command line argument.

New edl2segy program, which can be used to extract shot or receiver gathers from miniSEED files and save them as SEG-Y or Seismic Unix files.

New edlcut utility to cut out data samples from miniSEED files based on a given start time and a trace length.