jgeudens / graphviewer Goto Github PK
View Code? Open in Web Editor NEWLicense: GNU General Public License v3.0
License: GNU General Public License v3.0
Add a progress bar while loading files. Some files (>20 MB) take some time to load.
Windows indicate the application as "not responding". Maybe load in separate thread?
Wouldn't it be nice to represent the time as hh:mm:ss on the x-axis instead of autoscaling point numbers? It's a bit more readable :).
Import data fails when line is larger than 512 characters.
When the user has zoomed in on a graph then the dynamic reload function should not re-scale the graph when updating. The data points should be added, but the visible window should remain the same.
Hi,
I want to load a file with absolute date/time column. This seems not to be possible?
For example:
DateTime;Supply;Exhaust;T1;T2;T3
7/07/2017 16:00:01;3500;3501;28,2;-3,9;24,4
7/07/2017 16:10:01;3499;3500;28,3;-3,8;24,5
7/07/2017 16:20:01;3500;3500;28,4;-3,7;24,6
When a data file is parsed, all the graphs in the file are visible. Add some way to temporary hide some graphs.
Add option in main window or in the load file dialog.
Wouldn't it be nice to change the default values when opening a file to the first row for titles, the second row and first column as start for data. Also, it would be nice to set the minimum of these components to these values, as 0 is actually invalid.
When a csv file is dragged in to the application, the application should open the load file dialog with some options (file path, ...) filled in.
Now when you want to hide a curve, you must right click, go to the curve name and show/hide it. Wouldn't it be nice to use the legend?
Proposed workflow:
The dynamic reload feature needs to be reimplemented. Currently the dynamic reload feature does a full reload (graph clear, data structure read, data read, reset all settings).
Improvements:
Use timer so that the number of reloads per seconds is limited (prevent high cpu usage). Optimize reload so that only new data is read. Structure of data (number of columns) should not change between reloads. This should ensure that runtime settings (like hidden graphs) are remembered during dynamic reload.
Wouldn't it be nice to have a comment character/string?
GraphViewer should then ignore this line, but still read into memory for future use (for labels in graph).
The highlight of sample points is disabled when there are not enough pixels per sample point. The current implementation fails when the user zooms out too much.
It occurs when the lower bound of the x axis range is smaller than the lower bound of the graph or when the upper bound of the x-axis is greater than the upper bound of the graph.
Reading values of more than 8 bits with STMStudio can generate corrupt values because there is no synchronization between the target and STMStudio. We can detect/correct these values by comparing them with surrounding values. This correction should be optional, because it might potentially hide issues in the data.
The corrupt values are pretty recognizable:
Changing comment character doesn't update preview of parse result.
Let user choose colors of graphs
Options in right click:
Rename project to GraphViewer
Add option to bring graph to front.
And provide a menu item to turn it on/off. And don't show the marks when zoomed out to avoid screen clutter.
Add 2 command line arguments:
GraphViewer.exe [file] [skipDataLoadDialog]
Wouldn't it be nice to have two cursors to select a start and end time. Then the delta time is shown and maybe some default information like: minimum and maximum between the cursors, the average value, peak-peak value...
Wouldn't it be nice to set the x- and/or y-scaling relative to a point?
Examples:
p = last x value
scale to: [x-10s, x](s = seconds)p = custom value
scale to: [x, x+30min]
The minus sign is detected as comment character with negative time values.
The minus sign should only be detected as comment sign when there are 2 consecutive signs.
Wouldn't it be nice that a correct preset is selected automatically when selecting a file.
Add tag in presets.xml and if the keyword is found in the filename then that preset is selected.
Wouldn't it be nice to have support for "NaN" during import.
QCustomplot will show a gap for this point. This change would facilitate simpler comparing of two datasets.
The user will need to prepare a special csv file with 2 different datasets for viewing.
The reload menu item is a great improvement! But ๐, wouldn't it be nice to automatically reload a file when changed (after a question box of course, and maybe add something like a dynamic session which automatically reload without a question box)?
Wouldn't it be nice if the legend is draggable?
Some tab indexes aren't correct in the Load Data File dialog.
Use first data row to determine columns and give graphs generic names.
Data is also multiplied when time is in seconds
Sometimes it is better to show the raw time data instead of the hh::mm:ss,milli notation.
Show preview of selected file in Choose file dialog.
Now when you want to bring a curve to front, you must right click, go to the curve name and click "Bring to front". Wouldn't it be nice to use the legend?
Proposed workflow:
On startup, read xml file with the several set of parse settings. Show these custom parse settings sets in dropdown when choosing file.
Wouldn't it be nice to add labels in graphs. This way you can split your graph in sections and explain what you see in that section.
With the attached file, the "Show value tooltip" shows the correct values. But when "Time data in milliseconds" is turned off, the tooltip shows strange values.
Modbusscope.zip
The x axis data is very small (ยตs) when viewing data from a oscilloscoop. Graphviewer should detect these small values and change the x axis scale accordingly.
It should also change the labels.
Windows Explorer is blocked when a log file is opened by dragging the file in GraphViewer. This block is caused by the "load file" dialog.
Solution is to generate a signal on the dragdrop event and don't handle file settings in the dragdrop event
The meaning of the checkbox "Timedata in millisecond" is not clear.
Current functionality:
Internal data is always in milliseconds.
Reported by @lsteynen
Wouldn't it be nice that the legend can be hidden.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.