docassemble.InterviewStats 0.2__tar.gz → 0.5.3__tar.gz

docassemble_interviewstats-0.5.3/MANIFEST.in

@@ -0,0 +1,18 @@
+include README.md
+graft docassemble/InterviewStats/data
+recursive-exclude * *.egg-info
+recursive-exclude .git *
+recursive-exclude venv *
+recursive-exclude .github *
+recursive-exclude .pytest_cache *
+recursive-exclude .vscode *
+recursive-exclude build *
+recursive-exclude dist *
+recursive-exclude * __pycache__
+recursive-exclude * *.pyc
+recursive-exclude * *.pyo
+recursive-exclude * *.orig
+recursive-exclude * *~
+recursive-exclude * *.bak
+recursive-exclude * *.swp
+

{docassemble.InterviewStats-0.2/docassemble.InterviewStats.egg-info → docassemble_interviewstats-0.5.3}/PKG-INFO

@@ -1,17 +1,24 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.4
 Name: docassemble.InterviewStats
-Version: 0.2
-Summary: A docassemble extension, to view stats from other interviews
-Home-page: https://courtformsonline.org/about/
-Author: Bryce Willey, Quinten Steenhuis
-Author-email: bwilley@suffolk.com
-License: The MIT License (MIT)
-Platform: UNKNOWN
+Version: 0.5.3
+Summary: A docassemble interview to view statistics from other saved interview responses.
+Author-email: Suffolk Legal Innovation and Technology Lab <litlab@suffolk.edu>
+License-Expression: MIT
+Project-URL: GitHub, https://github.com/SuffolkLITLab/docassemble-InterviewStats
 Description-Content-Type: text/markdown
 License-File: LICENSE
+Requires-Dist: bokeh>=2.4.2
+Requires-Dist: cenpy>=1.0.0.post4
+Requires-Dist: geopandas>=0.9.0
+Requires-Dist: numpy>=1.0.4
+Requires-Dist: pandas>=1.4.2
+Requires-Dist: requests>=2.27.1
+Dynamic: license-file
 
 # docassemble.InterviewStats
 
+[![PyPI version](https://badge.fury.io/py/docassemble.InterviewStats.svg)](https://badge.fury.io/py/docassemble.InterviewStats)
+
 A docassemble interview that lets you view statistics from other saved interview responses. 
 
 ## Getting Started
@@ -71,5 +78,3 @@ If you have specific feature requests, feel free to open an issue or make a PR!
 * Quinten Steenhuis
 * Bryce Willey
 
-
-

{docassemble.InterviewStats-0.2 → docassemble_interviewstats-0.5.3}/README.md

@@ -1,5 +1,7 @@
 # docassemble.InterviewStats
 
+[![PyPI version](https://badge.fury.io/py/docassemble.InterviewStats.svg)](https://badge.fury.io/py/docassemble.InterviewStats)
+
 A docassemble interview that lets you view statistics from other saved interview responses. 
 
 ## Getting Started

docassemble_interviewstats-0.5.3/docassemble/InterviewStats/__init__.py

@@ -0,0 +1 @@
+__version__ = '0.5.3'

docassemble_interviewstats-0.5.3/docassemble/InterviewStats/data/questions/data_report.yml

@@ -0,0 +1,405 @@
+---
+metadata:
+  title: Interview data reporting
+  short title: Data reporting
+  temporary session: True
+  required privileges: 
+    - admin
+---
+features:
+  question help button: True
+  #question back button: True
+  css: 
+    - styles.css
+---
+default screen parts: 
+  footer: |
+    [:sync-alt: Restart](${ url_of('restart') } )
+---
+modules:
+  - .snapshot_statistics
+  - .data_report
+---
+objects:
+  - xlsx_output: DAFile.using(filename="stats.xlsx")
+#--------------------------------------------------------------------------
+#	Main logic
+#--------------------------------------------------------------------------
+---
+id: interview order
+mandatory: True
+code: |
+  filename # Choose interview for data report
+  simple_summary # Allow date filtering
+  excel_formatted_data # Prepare data for graph and excel   
+  final # Display download and filter options    
+#--------------------------------------------------------------------------
+#	Question screens
+#--------------------------------------------------------------------------
+---
+question: |
+  Data Reporting
+subquestion: |
+  **Prerequisites**: 
+  
+  1. You have a survey-type interview (i.e. interviews that only collect answers but don't have a download screen);  
+  1. you are using the `save_input_data()` function in ALToolbox to capture the answers for data reporting purposes. ([See a demo](https://apps-dev.suffolklitlab.org/interview?i=docassemble.ALToolbox%3Adata%2Fquestions%2Fsave_input_data_demo.yml#page1) if you haven't used it before);  
+  1. your interview has been run successfully at least once.
+  
+  You can then use this `data reporting` feature to pull your data records and make some quick on-screen data reports. In addition, you can download a complete data dump of the saved data records into an Excel file, and use it to create charts, review free typed answers etc..
+  
+  Select an interview to see data reports:
+fields:
+  - Filename: filename 
+    datatype: combobox
+    code: |      
+      [{interview.get('filename'): interview.get('title')} for interview in interview_menu()] or get_filenames()
+---
+question: |
+  Quick Summary for ${ interview_title }
+subquestion: |
+  * Total sessions: ${ summary_stats[0] }
+  * First session date: ${ format_date(summary_stats[1], format='MM/dd/yyyy') }
+  * Latest session date: ${ format_date(summary_stats[2], format='MM/dd/yyyy') }
+  
+  % if summary_stats[0] > 10000:
+  You have a lot of sessions; some operations might take a long time.
+  % endif  
+  ---
+  Do you want to filter your data records by date?
+fields:      
+  - no label: filter_date
+    datatype: yesnoradio
+    default: False
+    required: False
+  - Find submissions that are: date_op
+    datatype: dropdown
+    show if: filter_date
+    code: |
+      [{'gt': 'after'},
+       {'lt': 'before'},
+       {'eq': 'on'},
+       {'between': 'between'}]
+  - this date: begin_date
+    datatype: date
+    show if: filter_date
+  - and this date: end_date
+    datatype: date
+    js show if: |         
+      val('date_op') == 'between'
+continue button field: simple_summary
+---
+id: final screen
+event: final
+question: |
+  ${interview_title} - Data Reports
+subquestion: |
+  Total submissions: ${ len(stats) }
+  
+  % if filter_date:  
+    Date filter: Records dated ${phrase(date_op)} ${begin_date}
+    % if phrase(date_op) == 'between':
+      and ${end_date}
+    % endif
+    <br>Filtered submissions: ${ record_count }
+  % endif
+    
+  % if have_multi_vars:
+  #### Multi-valued questions
+  The following fields are from multi-valued questions. Select one to see its table:<br>
+    % for col in col_headings_multi:  
+      [${col}](${ url_action('show_filtered_data_multi', filter_by = col)  })
+    % endfor
+  % endif
+  
+  % if have_single_vars:  
+  #### Single-valued questions (excluding text/area/numberic fields) 
+  The following fields are from single-valued questions, except those whose answer lack statistic value, such as user-typed answer or numeric input. 
+  
+  Select a field to see its table:<br>
+    % for col in col_headings_single:  
+      [${col}](${ url_action('show_filtered_data_single', filter_by = col)  })
+    % endfor
+  % endif
+  
+  #### All questions (excluding text/area/number fields)
+  [Click here](${ url_action('display_filtered_data_all')  }) to see all the question tables on one screen.
+
+  #### Download an excel file:
+  To extract complete data including text/area/numeric fields, make an excel file:<br>
+  [:file-excel: Make Excel Download](${ url_ask('generate_excel') })
+---
+id: single-valued filtered screen
+question: |
+  Data Report - Single value questions
+subquestion: |
+  **Number of submissions**: ${ record_count }<br>
+  **Variable Name**: ${ my_col_single }<br> 
+  **Table Content**: Answer count / % of total submission<br>
+  :bell: Total percentage should be about 100% (±0.5% due to rounding).
+
+  ${ df_single[my_col_single].rename_axis("Answer").to_frame().query('index=='+my_index_single).to_html(classes=('table','table-striped')) }  
+continue button field: display_filtered_data_single
+continue button label: Back
+---  
+id: multi-valued filtered screen
+question: |
+  Data Report - Multi-choices questions
+subquestion: |
+  **Number of submissions**: ${ record_count }<br>
+  **Variable Name**: ${ my_col_multi }<br>
+  **Table Content**: Answer count / % of total submission<br>
+  :bell: Total percentage can be over or under 100%.
+  
+  ${ df_multi[my_col_multi].rename_axis("Answer").to_frame().query('index=='+my_index_multi).to_html(classes=('table','table-striped')) }    
+continue button field: display_filtered_data_multi
+continue button label: Back
+---  
+id: all_variables screen
+question: |
+  Data Report - All questions
+subquestion: |  
+  Number of submissions: ${ record_count }<br>
+  **Table Content**: Answer count / % of total submission<br>
+  :bell: For multi-valued questions, the total percentage can be over or under 100%; for single-valued questions, the total percentage should be about 100% (±0.5% due to rounding).
+  
+  #### Multi-value questions
+  % for fld in df_multi:
+    ${ df_multi[fld].rename_axis("Answer").to_frame().query('index=='+str(answer_list_multi[col_headings_multi.index(fld)])).to_html(classes=('table','table-striped')) }
+  % endfor  
+  
+  #### Single-value questions
+  % for fld in df_single:  
+    ${ df_single[fld].rename_axis("Answer").to_frame().query('index=='+str((answer_list_single[col_headings_single.index(fld)]))).to_html(classes=('table','table-striped')) }  
+  % endfor 
+continue button field: display_filtered_data_all
+continue button label: Back
+#--------------------------------------------------------------------------
+#	Data extract code blocks and screens
+#--------------------------------------------------------------------------
+---
+code: |
+  summary_stats = get_summary_stats(filename)
+---
+# Extract full data from database into stats
+code: |
+  # Get stats using filename
+  if filename is not None:    
+    if 'data/questions' in filename or 'playground' in filename:      
+      stats = get_stats(filename)       
+    else:
+      stats = get_stats(filename.replace(':',':data/questions/'))    
+  else:    
+    no_interview_stats 
+  
+  # Check if stats has data
+  if not stats:
+    no_interview_stats
+  else:      
+    # Check if stats is data-reporting ready - it must have a "title" column
+    if 'title' not in stats[0].keys():
+      no_reportable_data
+---
+event: no_interview_stats
+question: Interview has no data!
+subquestion: |  
+  The interview you selected is <br>
+  ${interview_title}.
+  
+  This interview doesn't have any data. Try another one.  
+buttons:
+  - Restart: restart
+---
+event: no_reportable_data
+question: Interview has no reportable data, you should use Stats
+subquestion: | 
+  The interview you selected is <br>
+  ${interview_title}.
+  
+  This interview contains data not suitable for Data Reporting. You should use Stats.
+buttons:
+  - Restart: restart
+---
+code: |  
+  interview_title = next(iter([interview.get('title') for interview in interview_menu() if interview.get('filename') == filename]), '(interview)')
+  if stats and 'title' in stats[0].keys():
+    interview_title = stats[0]['title']   
+---
+# Filter records in stats according to the filter
+code: |    
+  filtered_stats = list()
+  
+  # Build filtered_stats if filter_date is selected
+  if filter_date:
+    filter_begin = format_date(begin_date, format='MM/dd/yyyy')
+    if date_op == 'between':
+      filter_end = format_date(end_date, format='MM/dd/yyyy')    
+    for record in stats:
+      record_date = format_date(record['modtime'], format='MM/dd/yyyy')    
+      if (date_op == 'eq' and record_date == filter_begin) or (date_op == 'gt' and record_date > filter_begin) or (date_op == 'lt' and record_date < filter_begin) or (date_op == 'between' and record_date >= filter_begin and record_date < filter_end):
+        filtered_stats.append(record) 
+  
+  # In case a wrong date/date_op pair is selected
+  if filter_date and not filtered_stats:
+    no_filtered_stats
+---
+event: no_filtered_stats
+question: No records in this period
+subquestion: |
+  % if date_op == 'between':
+  You selected records "${phrase(date_op)} ${begin_date} and ${end_date}". 
+  % else:
+  You selected records "${phrase(date_op)} ${begin_date}".
+  % endif
+  There aren't any records fit your selection. 
+  
+  Try another date.
+back button: simple_summary
+---
+need: filtered_stats
+code: |  
+  # Process the filtered data records
+  columns = get_columns(stats)  
+  if filtered_stats:
+    table_data = processing_data(raw_data = filtered_stats, headings = columns) 
+    record_count = len(filtered_stats)
+  else:
+    record_count = len(stats)  
+    table_data = processing_data(raw_data = stats, headings = columns) 
+  
+  # Extract data for multi-valued questions
+  have_multi_vars = False
+  if table_data[0]:
+    have_multi_vars = True
+    table_data_multi = table_data[0]
+    col_headings_multi = list(table_data[0].keys())
+    answer_list_multi = table_data[1]
+    
+  # Extract data for single-valued questions
+  have_single_vars = False
+  if table_data[2]:
+    have_single_vars = True
+    table_data_single = table_data[2]
+    col_headings_single = list(table_data[2].keys())  
+    # Extract possible answers for single-valued questions  
+    answer_list_single = table_data[3]
+#--------------------------------------------------------------------------
+#	Pandas related code blocks
+#--------------------------------------------------------------------------
+---
+code: |
+  import pandas as pd
+---
+code: |  
+  excel_formatted_data = pd.DataFrame()
+  df_single = pd.DataFrame()
+  df_multi = pd.DataFrame()
+  stats_size = 0
+  if stats is not None:
+    excel_formatted_data = pd.DataFrame(stats)    
+    stats_size = excel_formatted_data.shape[0]
+    if have_single_vars:
+      df_single = pd.DataFrame(table_data_single)
+    if have_multi_vars: 
+      df_multi = pd.DataFrame(table_data_multi)
+---
+event: show_filtered_data_single
+code: |
+  # Receive user chosen field
+  my_col_single = action_argument('filter_by')  
+  
+  # Find the index of my_col_single, and use it to set my_index_single
+  my_row_single = col_headings_single.index(my_col_single)  
+  my_index_single = str(answer_list_single[my_row_single])
+  
+  # Display single-valued fields table screen
+  force_ask('display_filtered_data_single')
+---
+event: show_filtered_data_multi
+code: | 
+  # Receive user chosen field
+  my_col_multi = action_argument('filter_by')
+  
+  # Find the index of my_col_multi, and use it to set my_index_multi
+  my_row_multi = col_headings_multi.index(my_col_multi)  
+  my_index_multi = str(answer_list_multi[my_row_multi])
+  
+  # Display multi-valued fields table screen
+  force_ask('display_filtered_data_multi')  
+---
+code: |   
+  row_single = col_headings_single.index(fld)  
+  index_single = str(answer_list_single[row_single])  
+  
+  row_multi = col_headings_multi.index(fld)  
+  index_multi = str(answer_list_multi[row_multi])
+#--------------------------------------------------------------------------
+#	Excel related code blocks and screens
+#--------------------------------------------------------------------------
+---
+event: generate_excel
+code: |  
+  if stats_size > 10000:
+    if make_excel_task.ready():
+      if make_excel_task.failed():
+        show_failed_screen
+      show_excel_download
+    else:
+      waiting_screen
+  else:
+    proper_name = space_to_underscore(filename) + "_stats"
+    xlsx_output.initialize(filename=proper_name + ".xlsx")
+    writer = pd.ExcelWriter(xlsx_output.path(),
+      engine='xlsxwriter',
+      engine_kwargs={'options': {'remove_timezone': True}})
+    excel_formatted_data.to_excel(writer, index=False)
+    writer.close()
+    xlsx_output.commit()
+    del writer
+    show_excel_download
+---
+event: show_excel_download
+question: |
+  Excel Download
+subquestion: |
+  [:file-excel: Download](${ xlsx_output.url_for() })
+---
+code: |
+  make_excel_task = background_action('excel_task', data=excel_formatted_data)
+---
+event: excel_task
+code: |
+  bg_data = action_argument('data')
+  proper_name = space_to_underscore(filename) + "_stats"
+  xlsx_output.initialize(filename=proper_name + ".xlsx")
+  writer = pd.ExcelWriter(xlsx_output.path(),
+    engine='xlsxwriter',
+    engine_kwargs={'options': {'remove_timezone': True}})
+  bg_data.to_excel(writer, index=False)
+  writer.close()
+  xlsx_output.commit()
+  del writer
+  background_response_action('excel_resp', the_excel=xlsx_output) 
+---
+event: waiting_screen
+id: waiting screen
+question: |
+  We're generating your excel sheet.
+subquestion: |
+  This may take a few minutes, so leave this tab open. Thanks for waiting.
+reload: True
+---
+event: show_failed_screen
+question: |
+  Sorry, something went wrong
+subquestion: |
+  We had some trouble trying to assemble your excel sheet.
+
+  ${ make_excel_task.result() }
+buttons:
+  - Restart: restart
+---
+event: excel_resp
+code: |
+  xlsx_output = action_argument('the_excel')
+  background_response()