Jungfraujoch/rugnux/ReportDocument.h

// SPDX-FileCopyrightText: 2026 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
// SPDX-License-Identifier: GPL-3.0-only

#pragma once

#include <cstdint>
#include <string>
#include <vector>

// One populated result object, rendered more than once.
//
// The report used to be printed as the writer walked the results, which fixed the order of the file
// to the order of the pipeline and made a summary at the top impossible - the warnings that decide
// the verdict are produced by the sections the verdict has to sit above. BuildReportDocument()
// therefore populates this document, and a renderer walks it afterwards: the text renderer in
// ResultReport.cpp reads ONLY from here and prints nothing on its own. A second rendering is a
// second walk over the same document, never a second hand-maintained emitter.

// The TYPE of a pathology, from a closed vocabulary a consumer can switch on, carried beside the
// open free-text description that says what actually happened to this crystal. One code per warning
// PRODUCER: a code appears in PATHOLOGY_FLAGS if and only if its warning fired, so lighting an
// indicator is one grep and needs no prose parsing. Add a producer, add a code here.
namespace PathologyCode {
    constexpr const char *NO_LATTICE = "NO_LATTICE";                     // nothing indexed
    constexpr const char *INDEXING_AMBIGUITY = "INDEXING_AMBIGUITY";     // alternative indexing unresolved
    constexpr const char *SYMMETRY_AMBIGUITY = "SYMMETRY_AMBIGUITY";     // the data do not pick a point group
    constexpr const char *CENTERING_UNTESTED = "CENTERING_UNTESTED";     // centring from the metric only
    constexpr const char *UNUSABLE_MERGE = "UNUSABLE_MERGE";             // the merged data carry no signal
    constexpr const char *LOW_COMPLETENESS = "LOW_COMPLETENESS";         // too little of reciprocal space
    constexpr const char *SWEEP_GAPS = "SWEEP_GAPS";                     // ranges that delivered much less
    constexpr const char *GONIO_SCALE = "GONIO_SCALE";                   // the stage rotation is mis-scaled
    constexpr const char *SPINDLE_CAP = "SPINDLE_CAP";                   // the mounting cost unique data
    constexpr const char *ANISOTROPY = "ANISOTROPY";                     // strongly direction-dependent
    constexpr const char *TWINNING = "TWINNING";                         // twinning indicated
    constexpr const char *MODEL_HAND = "MODEL_HAND";                     // data and model in opposite hands
    constexpr const char *MODEL_NOT_VALIDATED = "MODEL_NOT_VALIDATED";   // a model was given, not usable
    constexpr const char *REFERENCE_MISMATCH = "REFERENCE_MISMATCH";     // the reference MTZ does not match; free flags not inherited
    constexpr const char *CANCELLED = "CANCELLED";                       // the run did not finish
    constexpr const char *RESOLUTION_FIT = "RESOLUTION_FIT";             // the CC1/2 fall-off is not a fall-off
    constexpr const char *FLIGHT_PATH = "FLIGHT_PATH";                   // the assumed flight-path medium matters here
    constexpr const char *PSEUDO_TRANSLATION = "PSEUDO_TRANSLATION";     // translational pseudo-symmetry indicated
    // Both warnings under this code say the same thing to a consumer - a translation the cell in use
    // does not declare is present in the data, so the centring or the cell is wrong - and differ only
    // in how exact it is. ANISOTROPY and RESOLUTION_FIT are already shared this way.
    constexpr const char *LATTICE_TRANSLATION = "LATTICE_TRANSLATION";   // a translation the cell does not declare
    constexpr const char *HARMONIC_CONTAMINATION = "HARMONIC_CONTAMINATION"; // the beam carries a higher harmonic
    constexpr const char *SUPERCELL_POSSIBLE = "SUPERCELL_POSSIBLE";     // Bragg-like intensity at half-integer positions
    constexpr const char *MULTIPLE_LATTICES = "MULTIPLE_LATTICES";       // a second lattice carries much of the spot intensity
    constexpr const char *GEOMETRY_NOT_CONVERGED = "GEOMETRY_NOT_CONVERGED"; // the geometry walk ran out of rounds
    constexpr const char *SCALING_NOT_CONVERGED = "SCALING_NOT_CONVERGED";   // the per-frame scales hit their cap still moving
    constexpr const char *POWDER_RINGS = "POWDER_RINGS";                 // non-ice powder rings hold many spots
    constexpr const char *ICE_RINGS = "ICE_RINGS";                       // hexagonal-ice rings hold many spots
}

// A closed type plus an open description. The type is what machinery switches on; the text is what a
// person reads, and it is free-form on purpose - the vocabulary must not have to grow a code every
// time a sentence gains a number.
struct ReportWarning {
    std::string code;
    std::string text;
    // Advisories about the pipeline's own internals rather than about this crystal. They still fire,
    // still count in the developer report, and are kept out of the default one - a warning nobody can
    // act on is how users learn to skip the warnings.
    bool developer_only = false;
};

// A value with its type kept, so a rendering other than text does not have to parse the string back.
// `text` is always the exact spelling the text report prints, because that spelling is the interface.
struct ReportValue {
    enum class Type { Text, Integer, Real, Boolean, Enumerated };
    Type type = Type::Text;
    std::string text;
    double real = 0.0;
    int64_t integer = 0;
    bool boolean = false;
    std::vector<std::string> domain;   // Enumerated only: the closed vocabulary this value comes from
};

// Content that is a table in every rendering. The pre-formatted text lines and the typed cells carry
// the same content: the text renderer prints the former, another rendering reads the latter.
struct ReportTable {
    std::vector<std::string> columns;
    std::string text_header;
    std::vector<std::string> text_rows;
    std::vector<std::vector<ReportValue>> cells;
};

struct ReportEntry {
    enum class Kind { Key, Prose, Blank, Table };
    Kind kind = Kind::Key;
    std::string key;         // Key
    ReportValue value;       // Key
    std::string prose;       // Prose: a paragraph, already line-wrapped for the text rendering
    ReportTable table;       // Table
    // Written only when the developer report was asked for. The document always holds everything;
    // which rendering shows what is the renderer's decision, so nothing is lost by building once.
    bool developer_only = false;
    // The other half of the same idea: a short paragraph that exists only because the long one it
    // summarises was moved out. Printing both would say the same thing twice.
    bool default_only = false;
};

struct ReportSection {
    std::string title;       // empty for the header block, which has no banner of its own
    std::vector<ReportEntry> entries;
    bool developer_only = false;
};

// The whole report as data. Sections in the order they are written; warnings and the verdict live
// here rather than in a section, because the SUMMARY section is composed from them after every other
// section has been built.
struct ReportDocument {
    std::vector<ReportSection> sections;
    std::vector<ReportWarning> warnings;
    std::string verdict;             // closed: OK | WARNINGS | UNUSABLE | FAILED
    std::string verdict_text;        // one to three sentences, free text
    std::vector<std::string> pathology_flags;   // PathologyCode::*, deduplicated, in first-seen order
};