ltc-paper-scripts 0.1.0__py3-none-any.whl

ltc_paper_scripts/README.md

@@ -0,0 +1,87 @@
+# Paper PDF build scripts - November 4, 2026
+
+Standalone snapshot of the approved 17-page paper exhibit builder. Copy this
+entire folder into Census. It does not import anything from the parent
+`ltc_code` package or the separate `paper/` folder.
+
+This folder is distributed as `ltc-paper-scripts` 0.1.0. It contains the paper
+scripts and assets only. Installing it has no declared dependencies and does
+not install rendering libraries. There is no `pyproject.toml`, `__init__.py`,
+`uv.lock`, or `.gitignore` in the download. The wheel includes only the required
+PyPI distribution metadata in addition to this folder.
+
+To locate the installed folder without an initialization file:
+
+```python
+import ltc_paper_scripts
+print(next(iter(ltc_paper_scripts.__path__)))
+```
+
+Copy the whole folder into your editable Census code directory. The normal
+command `python build_latex.py` uses the libraries already in that environment.
+The requirements file and Bash runner are optional conveniences for setting up
+an environment where dependency installation is permitted.
+
+## Run in Census
+
+Use Python 3.9 or newer, the Python libraries in `latex_requirements.txt`, and
+**pdfLaTeX** with the TeX packages listed in `latex_source/paper/preamble.tex`.
+The build requires no R, XeLaTeX, LuaLaTeX, shell escape, or internet access
+once those dependencies are installed. It compiles three pdfLaTeX passes.
+
+1. Supply your real results workbook. The included workbook combines existing
+   disclosed/example values and synthetic examples; it is not a new release.
+2. Edit `paths.py`: set `PENDING_INPUT`, the output location, and `PDFLATEX` if
+   the executable is not on PATH. Keep `PAPER_ROOT` at this folder, or adjust
+   it when using a shared `paths.py` in a separate code directory.
+3. Edit `input_mapping()` in `build_latex.py`: sheet/outcome/header names,
+   sample/specification names, the network roster, units, and benchmarks as
+   needed. Preserve the descriptive mapping keys on the left.
+4. From this folder, run:
+
+   ```sh
+   python build_latex.py
+   ```
+
+On a machine with `uv`, the equivalent command is:
+
+```sh
+uv run --no-project --with-requirements latex_requirements.txt python build_latex.py
+```
+
+`run_latex.sh` is an optional convenience for systems with Bash. Census does
+not need Bash or `uv` if Python and the required libraries are already available.
+
+The output is `output/charter_exhibits_latex.pdf`, accompanied by an audit JSON.
+The `output/latex_build/` directory contains the generated TeX and compile logs.
+There is no preview banner on the PDF pages.
+
+## Included exhibits and calculations
+
+- Figures 1-3, Tables 1-7, Figures A2-A4, and Tables A1-A4.
+- Excludes Figure 4, Figure A1, Tables A5-A7, and Table 1 Panel E.
+- All results default to the new workbook. Table results are red; figures
+  preserve their orange/teal colors.
+- Wage percent = `100 * wage coef / wage CCM`. Table 6 uses this same unrounded
+  percentage for every mediation denominator; there is no fixed 13.6 value.
+- Real inputs must contain the required results in the mapped units. This
+  builder consumes estimates, standard errors, control means, counts, and
+  summary statistics; it does not estimate regressions from microdata.
+
+## Files to keep together
+
+Keep every included Python file, `table_bindings.json`, the entire
+`latex_source/` tree, and `fonts/`. These are the core renderer and its assets.
+Only `paths.py` and `build_latex.py` need routine configuration. The example
+workbook can be replaced with real results. The old disclosed workbook is not
+required unless you explicitly opt into `reuse_disclosed`.
+
+The open DejaVu fonts and their license are bundled for portability. Installed
+Lucida Sans is preferred where available; otherwise typography uses DejaVu.
+Proprietary Office fonts are not included. The tick algorithm's license is in
+`TICK_ALGORITHM_LICENSE.txt`.
+
+Copied from `src/ltc_code/paper/` on November 4, 2026. Runtime files and templates
+match that approved builder; the README is specific to this standalone copy.
+The requirements file permits ReportLab 4.5 and newer. On Python 3.9 it selects
+ReportLab 4.x because ReportLab 5.1 requires Python 3.10 at runtime.

ltc_paper_scripts/TICK_ALGORITHM_LICENSE.txt

@@ -0,0 +1,26 @@
+Copyright (c) 2016, Hassan Kibirige
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+ * Redistributions of source code must retain the above copyright notice,
+   this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above copyright
+   notice, this list of conditions and the following disclaimer in the
+   documentation and/or other materials provided with the distribution.
+ * Neither the name of  nor the names of its contributors may be used to
+   endorse or promote products derived from this software without specific
+   prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+POSSIBILITY OF SUCH DAMAGE.

ltc_paper_scripts/build_latex.py

@@ -0,0 +1,293 @@
+# --- Import necessary packages ---
+import hashlib
+import json
+import re
+import shutil
+import subprocess
+import sys
+from pathlib import Path
+
+# Run from your code folder, where the shared paths.py lives.
+# Replace "." with that folder's absolute path if you run from elsewhere.
+sys.path.append(".")
+from paths import *
+from mappings import resolve_inputs
+from split_inputs import SplitInputs, build_tables
+from latex_figures import make_figures, FIGURE_EXHIBITS
+from exhibit_selection import select_exhibits
+from pypdf import PdfReader
+
+###########################################################
+# Edit workbook mappings here. Do not edit renderer or TeX files.
+###########################################################
+
+COLOR_PENDING = True  # Red pending table values; figures retain blue/orange.
+# Keep manuscript numbering even when exhibits are omitted.
+EXCLUDE_TABLES = ['lottery_sample_selection_table', 'cmo_details_table', 'intervention_calculation_table']
+EXCLUDE_FIGURES = ['Figure 4', 'Figure A.1']
+
+
+def input_mapping():
+    """Edit only the sheet names and variable names below for your Census file.
+
+    Paths are in paths.py. All results default to the pending workbook; reuse_disclosed opts into old results.
+    Keep the descriptive labels on the LEFT; edit the pair on the RIGHT.
+    Each pair is (sheet, variable). Use a quoted sheet name for any exception.
+    """
+    SUMMARY = 'summary_stats'
+    BALANCE = 'balance'
+    EFFECTS = 'TOT'
+    OLS = 'OLS'
+    FIRST_STAGE = 'first_stage'
+    ITT = 'ITT'
+    CONVERSIONS = 'mediation_conversions'
+    NETWORKS = 'TOT_by_cmo'
+    PREDICTIONS = 'predictions'  # Change to 'TOTE' if predictions live there.
+
+    inputs = {
+        # Table 1: summary statistics
+        'Summary: parent income': (SUMMARY, 'parent_income'),
+        'Summary: parent inc rank': (SUMMARY, 'parent_inc_rank'),
+        'Summary: female': (SUMMARY, 'female'),
+        'Summary: hispanic': (SUMMARY, 'hispanic'),
+        'Summary: black': (SUMMARY, 'black'),
+        'Summary: white': (SUMMARY, 'white'),
+        'Summary: entry age': (SUMMARY, 'entry_age'),
+        'Summary: enrollment year': (SUMMARY, 'enrollment_year'),
+        'Summary: enrollment years': (SUMMARY, 'enrollment_years'),
+        'Summary: Age 25 earnings': (SUMMARY, 'wages25'),
+        'Summary: 1098T college by 20': (SUMMARY, 'any1098T_by20'),
+        'Summary: College attendance by 20': (SUMMARY, 'att_any_by_age20'),
+        'Summary: employed25': (SUMMARY, 'employed25'),
+        'Summary: n ppl': (SUMMARY, 'n_ppl'),
+        # Table 2: balance
+        'Balance: black': (BALANCE, 'black'),
+        'Balance: hispanic': (BALANCE, 'hispanic'),
+        'Balance: white': (BALANCE, 'white'),
+        'Balance: female': (BALANCE, 'female'),
+        'Balance: parent inc rank': (BALANCE, 'parent_inc_rank'),
+        'Balance: med hhinc2000': (BALANCE, 'med_hhinc2000'),
+        'Balance: wages25 pre treatment': (BALANCE, 'wages25_pre_treatment'),
+        'Balance: linked tax': (BALANCE, 'linked_tax'),
+        'Balance: joint covariate test': (BALANCE, 'joint_covariates'),
+        # Effects shared across tables
+        'Age 25 household income': (EFFECTS, 'agi25'),
+        'Age 25 earnings': (EFFECTS, 'wages25'),
+        'Surrogate earnings at 31 by subgroup': (EFFECTS, 'wages31_firm_chars'),
+        '1098T college by 20': (EFFECTS, 'any1098T_by20'),
+        'BA completion by year 6': (EFFECTS, 'cmp_BA_byY6'),
+        'College attendance by 20': (EFFECTS, 'att_any_by_age20'),
+        'Two-year attendance by 20': (EFFECTS, 'att_2yr_by_age20'),
+        'Four-year attendance by 20': (EFFECTS, 'att_4yr_by_age20'),
+        'Elite college attendance by 20': (EFFECTS, 'att_elite_by_age20'),
+        'Elite BA completion by year 6': (EFFECTS, 'cmp_elite_byY6'),
+        'Firm selective BA rate': (EFFECTS, 'firm19_selective_ba'),
+        'Firm worker earnings': (EFFECTS, 'firm19_wages30_40'),
+        'Firm retention': (EFFECTS, 'firm19_retention'),
+        'Age 25 spouse earnings': (EFFECTS, 'spouse_wages25'),
+        'Married at 25': (EFFECTS, 'married25'),
+        'Any completion by year 6': (EFFECTS, 'cmp_any_byY6'),
+        'AA completion by year 6': (EFFECTS, 'cmp_AA_byY6'),
+        'Predicted any completion': (EFFECTS, 'pred_cmp_any'),
+        'Predicted four-year completion': (EFFECTS, 'pred_cmp_4yr'),
+        'Predicted two-year completion': (EFFECTS, 'pred_cmp_2yr'),
+        'Selective BA completion by year 6': (EFFECTS, 'cmp_selective_byY6'),
+        'Figure A2: college at each age': (EFFECTS, 'att_any_age{age}'),
+        'Figure A4: firm surrogate earnings': (EFFECTS, 'wages{age}_firm_chars'),
+        # Table A1: observational OLS comparison mean
+        'OLS: Age 25 earnings': (OLS, 'wages25'),
+        # Table A2: predicted earnings
+        'Predicted earnings above $20,000': (PREDICTIONS, 'wages25_over_20000'),
+        'Predicted earnings above $50,000': (PREDICTIONS, 'wages25_over_50000'),
+        'Predicted earnings above $100,000': (PREDICTIONS, 'wages25_over_100000'),
+        # Previously disclosed results and newly included figures
+        'First stage: charter enrollment': (FIRST_STAGE, 'enrolled'),
+        'ITT: Age 25 earnings': (ITT, 'wages25'),
+        'Peers: parent income': (OLS, 'peer_parent_income'),
+        'Teen birth by age 20': (EFFECTS, 'teenbirth_claim20'),
+        'Actual earnings above $0': (EFFECTS, 'wages25_over_0'),
+        'Actual earnings above $20,000': (EFFECTS, 'wages25_over_20000'),
+        'Actual earnings above $50,000': (EFFECTS, 'wages25_over_50000'),
+        'Actual earnings above $100,000': (EFFECTS, 'wages25_over_100000'),
+        'Mediation: BA earnings conversion': (CONVERSIONS, 'cmp_BA_byY6'),
+        'Mediation: selective BA earnings conversion': (CONVERSIONS, 'cmp_selective_byY6'),
+        'Mediation: elite BA earnings conversion': (CONVERSIONS, 'cmp_elite_byY6'),
+        'Figures 1/A4: observed earnings at each age': (EFFECTS, 'wages{age}'),
+        'Figure 1: combined surrogate earnings': (EFFECTS, 'wages{age}_ein_fes'),
+        'Figure 3: 2yr attendance at age 20': (EFFECTS, 'att_2yr_age20'),
+        'Figure 3: 4yr attendance at age 20': (EFFECTS, 'att_4yr_age20'),
+        'Figure 3: elite attendance at age 20': (EFFECTS, 'att_elite_age20'),
+        'Figure 3: AA completion by age 24': (EFFECTS, 'cmp_AA_by_age24'),
+        'Figure 3: BA completion by age 24': (EFFECTS, 'cmp_BA_by_age24'),
+        'Figure 3: elite completion by age 24': (EFFECTS, 'cmp_elite_by_age24'),
+        'Figure A3: network earnings': (NETWORKS, 'wages25'),
+    }
+    # COLUMN HEADERS: left = what the code needs; right = your Excel header.
+    # Change independently for each sheet. E.g. SUMMARY 'variable': 'depvar',
+    # 'mean': 'average', 'count': 'N_ppl'. Add any other field rename here.
+    columns = {
+        SUMMARY: {'sample': 'sample', 'variable': 'variable',
+                  'mean': 'mean', 'sd': 'sd', 'count': 'count'},
+        BALANCE: {'sample': 'sample', 'depvar': 'depvar', 'control_mean': 'control_mean',
+                  'coef': 'coef', 'se': 'se', 'N_ppl': 'N_ppl',
+                  'f_stat': 'f_stat', 'p_value': 'p_value'},
+        EFFECTS: {'sample': 'sample', 'depvar': 'depvar', 'spec': 'spec',
+                  'coef': 'coef', 'se': 'se', 'ccm': 'ccm', 'N_ppl': 'N_ppl'},
+        FIRST_STAGE: {'sample':'sample', 'depvar':'depvar', 'spec':'spec',
+                      'coef':'coef', 'se':'se', 'N_ppl':'N_ppl', 'control_mean':'control_mean'},
+        ITT: {'sample':'sample', 'depvar':'depvar', 'spec':'spec',
+              'coef':'coef', 'se':'se', 'N_ppl':'N_ppl', 'control_mean':'control_mean'},
+        CONVERSIONS: {'depvar':'depvar', 'earnings_gain':'earnings_gain'},
+        NETWORKS: {'sample':'sample', 'depvar':'depvar', 'spec':'spec', 'coef':'coef', 'se':'se'},
+        OLS: {'sample': 'sample', 'depvar': 'depvar', 'spec': 'spec',
+              'comparison_mean': 'comparison_mean', 'coef':'coef', 'se':'se', 'N_ppl':'N_ppl'},
+        PREDICTIONS: {'sample': 'sample', 'depvar': 'depvar', 'spec': 'spec', 'coef': 'coef'},
+    }
+    # SAMPLE VALUES: left = the table's sample; right = its name in your workbook.
+    # 'tract_sample' is the tract comparison sample; 'national' is the national sample.
+    # 'aligned_20plus' is the college sample aligned to 1098T coverage.
+    samples = {
+        'figure3_attendance': 'overall_20plus',
+        'figure3_completion': 'overall_25plus',
+        'aligned_20plus': 'aligned_20plus',
+        'black_20plus': 'black_20plus',
+        'black_25plus': 'black_25plus',
+        'entry_high_school_20plus': 'entry_high_school_20plus',
+        'entry_high_school_25plus': 'entry_high_school_25plus',
+        'entry_middle_school_20plus': 'entry_middle_school_20plus',
+        'entry_middle_school_25plus': 'entry_middle_school_25plus',
+        'ever_college_25plus': 'ever_college_25plus',
+        'female_20plus': 'female_20plus',
+        'female_25plus': 'female_25plus',
+        'hispanic_20plus': 'hispanic_20plus',
+        'hispanic_25plus': 'hispanic_25plus',
+        'male_20plus': 'male_20plus',
+        'male_25plus': 'male_25plus',
+        'national': 'national',
+        'neighborhood_income_high_20plus': 'neighborhood_income_high_20plus',
+        'neighborhood_income_high_25plus': 'neighborhood_income_high_25plus',
+        'neighborhood_income_low_20plus': 'neighborhood_income_low_20plus',
+        'neighborhood_income_low_25plus': 'neighborhood_income_low_25plus',
+        'never_college_25plus': 'never_college_25plus',
+        'overall_20plus': 'overall_20plus',
+        'overall_25plus': 'overall_25plus',
+        'parent_income_high_20plus': 'parent_income_high_20plus',
+        'parent_income_high_25plus': 'parent_income_high_25plus',
+        'parent_income_low_20plus': 'parent_income_low_20plus',
+        'parent_income_low_25plus': 'parent_income_low_25plus',
+        'tract_sample': 'tract_sample',
+    }
+    # If an alternative result is identified by SAMPLE instead of SPEC, set it here.
+    # Example: {'Age 25 earnings': {'initial_offer': 'your_initial_offer_sample'}}
+    # This replaces that result's sample and removes its spec filter.
+    sample_instead_of_spec = {}
+    # Missing spec columns are automatic. Only rename existing spec values here.
+    specs = {}  # Example: {'baseline': 'none'}
+    # Every result is new/red by default. To reuse one result in black, add:
+    # reuse_disclosed = {'ITT: Age 25 earnings': ('ITT', 'wages25')}
+    # Reused results keep the original column/sample/spec names. New-name maps
+    # below apply only to PENDING_INPUT, so they cannot alter old-workbook lookups.
+    reuse_disclosed = {}
+    # Summary means/SDs: 1 for percentage points (0-100), 100 for proportions (0-1).
+    summary_percent_scale = 1
+    # Figure A3: display label -> sample value in the network-results sheet.
+    # Replace this example roster with your actual networks; do not include Overall.
+    networks = {
+        'Network A': 'network_a', 'Network B': 'network_b',
+        'Network C': 'network_c', 'Network D': 'network_d',
+        'Network E': 'network_e', 'Network F': 'network_f',
+    }
+    # External benchmarks used in the test-score mediation calculation.
+    benchmarks = {'credo_math_effect_per_year': 0.0974,
+                  'star_earnings_return_per_sd': 13.17, 'years_of_exposure': 3}
+    return resolve_inputs(inputs, columns=columns, samples=samples, specs=specs,
+                          sample_instead_of_spec=sample_instead_of_spec,
+                          reuse_disclosed=reuse_disclosed,
+                          summary_percent_scale=summary_percent_scale,
+                          networks=networks, benchmarks=benchmarks)
+
+
+###########################################################
+# Build the approved exhibit packet with pdfLaTeX directly.
+###########################################################
+
+def build_latex(mapping=None, disclosed_path=None, pending_path=None, output_path=None, build_dir=None):
+    mapping = input_mapping() if mapping is None else mapping
+    paths = {'disclosed': Path(disclosed_path or DISCLOSED_INPUT), 'pending': Path(pending_path or PENDING_INPUT)}
+    output_path = Path(output_path or OUTPUT_PDF)
+    build_dir = Path(build_dir or BUILD_DIR)
+    if build_dir.resolve() in {SOURCE_DIR.resolve(), PAPER_ROOT.resolve()}:
+        raise ValueError('BUILD_DIR must be separate from source files.')
+    bindings = json.loads(BINDINGS_FILE.read_text())
+    main, appendix, included, excluded, references = select_exhibits(SOURCE_DIR, EXCLUDE_TABLES, EXCLUDE_FIGURES)
+    data = SplitInputs(paths, mapping, color_pending=COLOR_PENDING)
+    table_names = {e['name'] for e in included if e['kind']=='table'}
+    # Validate before invoking TeX. Only selected exhibit inputs are required.
+    active_cells = {key:b for key,b in bindings['cells'].items() if key.split('.')[0] in table_names}
+    try:
+        for binding in active_cells.values(): data.render(binding)
+        build_dir.mkdir(parents=True, exist_ok=True)
+        shutil.copytree(SOURCE_DIR/'paper', build_dir/'paper', dirs_exist_ok=True)
+        # Write tables into a controlled build tree, never over inputs/templates.
+        build_tables(data, dict(bindings, cells=active_cells), SOURCE_DIR, build_dir)
+        figures = mapping['figures']
+        selected_figures = {e['name'] for e in included if e['kind']=='figure'}
+        required_figures = {key for key, exhibit in FIGURE_EXHIBITS.items() if exhibit in selected_figures}
+        if required_figures - figures.keys():
+            raise ValueError(f'Missing selected figure mappings: {sorted(required_figures - figures.keys())}')
+        figures = {k:v for k,v in figures.items() if FIGURE_EXHIBITS[k] in selected_figures}
+        figure_audit = make_figures(data, figures, build_dir/'outputs/figures')
+        # Table notes use the same live inputs as the corresponding numeric cells.
+        main = re.sub(r'@@([^@]+)@@', lambda m: data.render(active_cells[m[1]]), main)
+        appendix = re.sub(r'@@([^@]+)@@', lambda m: data.render(active_cells[m[1]]), appendix)
+        preview = any('preview' in str(row.get('source', '')).lower() or
+                      'synthetic' in str(row.get('source', '')).lower() or
+                      'hypothetical' in str(row.get('source', '')).lower()
+                      for row in data.records.values())
+        (build_dir/'paper/main-tables.tex').write_text(main)
+        (build_dir/'paper/appendix.tex').write_text(appendix)
+        wrapper = r'''\documentclass[11pt,english]{article}
+\makeatletter
+\def\input@path{{paper/}}
+\makeatother
+\input{preamble}
+\begin{document}
+\input{main-tables}
+\begin{appendices}
+\input{appendix}
+\end{appendices}
+\end{document}
+'''
+        (build_dir/'main.tex').write_text(wrapper)
+        engine = shutil.which(str(PDFLATEX))
+        if engine is None:
+            raise RuntimeError('pdfLaTeX was not found. Set PDFLATEX in paths.py to its executable path.')
+        command = [engine, '-interaction=nonstopmode', '-halt-on-error', '-file-line-error', '-no-shell-escape', 'main.tex']
+        with (build_dir/'compile-output.txt').open('w') as log:
+            for pass_number in range(1,4):
+                result = subprocess.run(command, cwd=build_dir, capture_output=True, text=True)
+                log.write(f'PASS {pass_number}\n'+result.stdout+'\n'+result.stderr)
+                if result.returncode:
+                    raise RuntimeError('pdfLaTeX failed. See '+str(build_dir/'compile-output.txt')+'\n'+(result.stdout+result.stderr)[-3500:])
+        pdf = PdfReader(build_dir/'main.pdf')
+        if len(pdf.pages) != len(included):
+            raise ValueError(f'Expected {len(included)} exhibit pages; got {len(pdf.pages)}.')
+        for page, exhibit in zip(pdf.pages, included):
+            if exhibit['title'].upper() not in page.extract_text():
+                raise ValueError('Missing or renumbered exhibit: '+exhibit['title'])
+        output_path.parent.mkdir(parents=True, exist_ok=True)
+        shutil.copyfile(build_dir/'main.pdf', output_path)
+        audit = {'engine':engine, 'pages':len(pdf.pages), 'included':[e['title'] for e in included],
+                 'preview':preview,
+                 'input_sha256':{role:hashlib.sha256(paths[role].read_bytes()).hexdigest() for role in data.books},
+                 'mapping':mapping, 'used':data.used, 'figures':figure_audit,
+                 'cells':{key:{'value':data.value(b), 'pending':data.is_pending(b)} for key,b in active_cells.items()}}
+        output_path.with_suffix('.audit.json').write_text(json.dumps(audit,indent=2)+'\n')
+        print(f'Created {output_path}: {len(pdf.pages)} pages with pdfLaTeX')
+        return audit
+    finally:
+        data.close()
+
+
+if __name__ == '__main__':
+    build_latex()

ltc_paper_scripts/exhibit_selection.py

@@ -0,0 +1,52 @@
+# --- Import necessary packages ---
+import re
+
+###########################################################
+# Select complete exhibits and retain their manuscript numbering
+###########################################################
+
+FIGURES = (
+    'appendix-credo-distributions',
+    'appendix-college-by-age-placeholder',
+    'appendix-cmo-leave-one-out',
+    'appendix-earnings-by-age-firm-chars',
+)
+FLOAT = re.compile(r'\\begin\{(table|figure)\}.*?\\end\{\1\}',re.S)
+
+
+def select_exhibits(source_root, exclude_tables, exclude_figures):
+    main=(source_root/'paper/tables-and-figures.tex').read_text()
+    # Retain the main figures; the catalog owns exclusions and manuscript numbering.
+    appendix=(source_root/'paper/appendix.tex').read_text()
+    catalog=[]
+    for section,text in [('main',main),('appendix',appendix)]:
+        counters={'table':0,'figure':0}
+        for match in FLOAT.finditer(text):
+            kind=match[1];counters[kind]+=1;block=match[0]
+            label=re.search(r'\\label\{([^}]+)\}',block)[1]
+            name=re.search(r'\\input\{[^}]*?/([^/}]+)\}',block)[1] if kind=='table' else label.removeprefix('fig:')
+            number=('A.' if section=='appendix' else '')+str(counters[kind])
+            catalog.append(dict(section=section,kind=kind,name=name,label=label,number=number,
+                                title=kind.title()+' '+number,counter=counters[kind]-1))
+    excluded=set()
+    for kind,requested in [('table',exclude_tables),('figure',exclude_figures)]:
+        if isinstance(requested,str):raise ValueError(f'EXCLUDE_{kind.upper()}S must be a list, not a string')
+        aliases={alias:e['name'] for e in catalog if e['kind']==kind for alias in [e['name'],e['title']]}
+        for name in requested:
+            if name not in aliases:raise ValueError(f'Unknown excluded {kind}: {name!r}. Choose from: {sorted(aliases)}')
+            excluded.add(aliases[name])
+    retained=[e for e in catalog if e['name'] not in excluded]
+    if not retained:raise ValueError('All exhibits are excluded. Include at least one table or figure.')
+    if excluded:
+        for section,text in [('main',main),('appendix',appendix)]:
+            entries=iter(e for e in catalog if e['section']==section)
+            def replace(match):
+                entry=next(entries)
+                if entry['name'] in excluded:return ''
+                return '\\setcounter{'+entry['kind']+'}{'+str(entry['counter'])+'}\n'+match[0]
+            filtered=FLOAT.sub(replace,text)
+            if section=='main':main=filtered
+            else:appendix=filtered
+    references='\n'.join('\\newlabel{'+e['label']+'}{{'+e['number']+'}{}{}{'+e['kind']+'.'+e['number']+'}{}}%'
+                         for e in catalog if e['name'] in excluded)
+    return main,appendix,retained,[e for e in catalog if e['name'] in excluded],references

ltc_paper_scripts/figure_ticks.py

@@ -0,0 +1,51 @@
+"""Dependency-free extended Wilkinson breaks, as used by scales::extended_breaks.
+
+Adapted from mizani.breaks.breaks_extended (Hassan Kibirige / Justin Talbot).
+See TICK_ALGORITHM_LICENSE.txt. Kept local to avoid numpy/pandas dependencies
+in the portable workbook-to-LaTeX builder.
+"""
+import math
+
+
+def extended_breaks(lo, hi):
+    if hi == lo:
+        return [lo]
+    span = hi-lo
+    best_score, best = -2, []
+    q_values = [1, 5, 2, 2.5, 4, 3]
+    j = 1
+    while True:
+        for index, q in enumerate(q_values):
+            sm = (5-index)/5 + 1-j
+            if .25*sm + .75 < best_score:
+                return [v for v in best if lo-1e-9 <= v <= hi+1e-9]
+            k = 2
+            while True:
+                dm = 2-(k-1)/4 if k >= 5 else 1
+                if .25*sm + .2 + .5*dm + .05 < best_score:
+                    break
+                z = math.ceil(math.log10(span/(k+1)/j/q))
+                while True:
+                    step = j*q*10**z
+                    cm = 1-((step*(k-1)-span)/2)**2/(.1*span)**2 if step*(k-1)>span else 1
+                    if .25*sm + .2*cm + .5*dm + .05 < best_score:
+                        break
+                    first = math.floor(hi/step)*j-(k-1)*j
+                    last = math.ceil(lo/step)*j
+                    if first > last:
+                        break
+                    for start in range(first,last+1):
+                        a = start*step/j
+                        b = a+step*(k-1)
+                        zero = int((a % step < 1e-10 or step-a % step < 1e-10) and a<=0<=b)
+                        simplicity = (5-index)/5 + zero-j
+                        coverage = 1-.5*((hi-b)**2+(lo-a)**2)/(.1*span)**2
+                        density_ratio = ((k-1)/(b-a))/(4/(max(b,hi)-min(a,lo)))
+                        density = 2-max(density_ratio,1/density_ratio)
+                        score = .25*simplicity+.2*coverage+.5*density+.05
+                        if score > best_score:
+                            best_score = score
+                            best = [a+i*step for i in range(k)]
+                    z += 1
+                k += 1
+        j += 1

ltc_paper_scripts/fonts/LICENSE.txt

@@ -0,0 +1,96 @@
+Fonts are (c) Bitstream (see below). DejaVu changes are in public domain. Glyphs imported from Arev fonts are (c) Tavmjung Bah (see below)
+
+Bitstream Vera Fonts Copyright
+------------------------------
+
+Copyright (c) 2003 by Bitstream, Inc. All Rights Reserved. Bitstream Vera is
+a trademark of Bitstream, Inc.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of the fonts accompanying this license ("Fonts") and associated
+documentation files (the "Font Software"), to reproduce and distribute the
+Font Software, including without limitation the rights to use, copy, merge,
+publish, distribute, and/or sell copies of the Font Software, and to permit
+persons to whom the Font Software is furnished to do so, subject to the
+following conditions:
+
+The above copyright and trademark notices and this permission notice shall
+be included in all copies of one or more of the Font Software typefaces.
+
+The Font Software may be modified, altered, or added to, and in particular
+the designs of glyphs or characters in the Fonts may be modified and
+additional glyphs or characters may be added to the Fonts, only if the fonts
+are renamed to names not containing either the words "Bitstream" or the word
+"Vera".
+
+This License becomes null and void to the extent applicable to Fonts or Font
+Software that has been modified and is distributed under the "Bitstream
+Vera" names.
+
+The Font Software may be sold as part of a larger software package but no
+copy of one or more of the Font Software typefaces may be sold by itself.
+
+THE FONT SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF COPYRIGHT, PATENT,
+TRADEMARK, OR OTHER RIGHT. IN NO EVENT SHALL BITSTREAM OR THE GNOME
+FOUNDATION BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, INCLUDING
+ANY GENERAL, SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES,
+WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF
+THE USE OR INABILITY TO USE THE FONT SOFTWARE OR FROM OTHER DEALINGS IN THE
+FONT SOFTWARE.
+
+Except as contained in this notice, the names of Gnome, the Gnome
+Foundation, and Bitstream Inc., shall not be used in advertising or
+otherwise to promote the sale, use or other dealings in this Font Software
+without prior written authorization from the Gnome Foundation or Bitstream
+Inc., respectively. For further information, contact: fonts at gnome dot
+org. 
+
+Arev Fonts Copyright
+------------------------------
+
+Copyright (c) 2006 by Tavmjong Bah. All Rights Reserved.
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of the fonts accompanying this license ("Fonts") and
+associated documentation files (the "Font Software"), to reproduce
+and distribute the modifications to the Bitstream Vera Font Software,
+including without limitation the rights to use, copy, merge, publish,
+distribute, and/or sell copies of the Font Software, and to permit
+persons to whom the Font Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright and trademark notices and this permission notice
+shall be included in all copies of one or more of the Font Software
+typefaces.
+
+The Font Software may be modified, altered, or added to, and in
+particular the designs of glyphs or characters in the Fonts may be
+modified and additional glyphs or characters may be added to the
+Fonts, only if the fonts are renamed to names not containing either
+the words "Tavmjong Bah" or the word "Arev".
+
+This License becomes null and void to the extent applicable to Fonts
+or Font Software that has been modified and is distributed under the 
+"Tavmjong Bah Arev" names.
+
+The Font Software may be sold as part of a larger software package but
+no copy of one or more of the Font Software typefaces may be sold by
+itself.
+
+THE FONT SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT
+OF COPYRIGHT, PATENT, TRADEMARK, OR OTHER RIGHT. IN NO EVENT SHALL
+TAVMJONG BAH BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+INCLUDING ANY GENERAL, SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL
+DAMAGES, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+FROM, OUT OF THE USE OR INABILITY TO USE THE FONT SOFTWARE OR FROM
+OTHER DEALINGS IN THE FONT SOFTWARE.
+
+Except as contained in this notice, the name of Tavmjong Bah shall not
+be used in advertising or otherwise to promote the sale, use or other
+dealings in this Font Software without prior written authorization
+from Tavmjong Bah. For further information, contact: tavmjong @ free
+. fr.