Skip to content

L
libcifpp
Commit 2792caec by Maarten L. Hekkelman
Options

renaming field and column to item

parent fb2b1e98
Showing with 499 additions and 234 deletions
include/cif++/condition.hpp
...@@ -39,17 +39,17 @@ ...@@ -39,17 +39,17 @@
* query you can use to find rows in a @ref cif::category * query you can use to find rows in a @ref cif::category
* *
* Conditions are created as standard C++ expressions. That means * Conditions are created as standard C++ expressions. That means
* you can use the standard comparison operators to compare field * you can use the standard comparison operators to compare item
* contents with a value and boolean operators to chain everything * contents with a value and boolean operators to chain everything
* together. * together.
* *
* To create a query that simply compares one field with one value: * To create a query that simply compares one item with one value:
* *
* @code {.cpp} * @code {.cpp}
* cif::condition c = cif::key("id") == 1; * cif::condition c = cif::key("id") == 1;
* @endcode * @endcode
* *
* That will find rows where the ID field contains the number 1. If * That will find rows where the ID item contains the number 1. If
* using cif::key is a bit too much typing, you can also write: * using cif::key is a bit too much typing, you can also write:
* *
* @code{.cpp} * @code{.cpp}
...@@ -64,7 +64,7 @@ ...@@ -64,7 +64,7 @@
* auto c3 = "id"_key == 1 or "id"_key == 2; * auto c3 = "id"_key == 1 or "id"_key == 2;
* @endcode * @endcode
* *
* There are some special values you can use. To find rows with field that * There are some special values you can use. To find rows with item that
* do not have a value: * do not have a value:
* *
* @code{.cpp} * @code{.cpp}
...@@ -83,7 +83,7 @@ ...@@ -83,7 +83,7 @@
* auto c6 = cif::all; * auto c6 = cif::all;
* @endcode * @endcode
* *
* And when you want to search for any column containing the value 'foo': * And when you want to search for any item containing the value 'foo':
* *
* @code{.cpp} * @code{.cpp}
* auto c7 = cif::any == "foo"; * auto c7 = cif::any == "foo";
...@@ -104,31 +104,40 @@ namespace cif ...@@ -104,31 +104,40 @@ namespace cif
/// we declare a function to access its contents /// we declare a function to access its contents
/** /**
* @brief Get the fields that can be used as key in conditions for a category * @brief Get the items that can be used as key in conditions for a category
* *
* @param cat The category whose fields to return * @param cat The category whose items to return
* @return iset The set of key field names * @return iset The set of key item names
*/ */
[[deprecated("use get_category_items instead")]]
iset get_category_fields(const category &cat); iset get_category_fields(const category &cat);
/** /**
* @brief Get the column index for column @a col in category @a cat * @brief Get the items that can be used as key in conditions for a category
*
* @param cat The category whose items to return
* @return iset The set of key field names
*/
iset get_category_items(const category &cat);
/**
* @brief Get the item index for item @a col in category @a cat
* *
* @param cat The category * @param cat The category
* @param col The name of the column * @param col The name of the item
* @return uint16_t The index * @return uint16_t The index
*/ */
uint16_t get_column_ix(const category &cat, std::string_view col); uint16_t get_item_ix(const category &cat, std::string_view col);
/** /**
* @brief Return whether the column @a col in category @a cat has a primitive type of *uchar* * @brief Return whether the item @a col in category @a cat has a primitive type of *uchar*
* *
* @param cat The category * @param cat The category
* @param col The column name * @param col The item name
* @return true If the primitive type is of type *uchar* * @return true If the primitive type is of type *uchar*
* @return false If the primitive type is not of type *uchar* * @return false If the primitive type is not of type *uchar*
*/ */
bool is_column_type_uchar(const category &cat, std::string_view col); bool is_item_type_uchar(const category &cat, std::string_view col);
// -------------------------------------------------------------------- // --------------------------------------------------------------------
// some more templates to be able to do querying // some more templates to be able to do querying
...@@ -219,7 +228,7 @@ class condition ...@@ -219,7 +228,7 @@ class condition
/** /**
* @brief Prepare the condition to be used on category @a c. This will * @brief Prepare the condition to be used on category @a c. This will
* take care of setting the correct indices for fields e.g. * take care of setting the correct indices for items e.g.
* *
* @param c The category this query should act upon * @param c The category this query should act upon
*/ */
...@@ -312,7 +321,7 @@ namespace detail ...@@ -312,7 +321,7 @@ namespace detail
condition_impl *prepare(const category &c) override condition_impl *prepare(const category &c) override
{ {
m_item_ix = get_column_ix(c, m_item_tag); m_item_ix = get_item_ix(c, m_item_tag);
return this; return this;
} }
...@@ -339,7 +348,7 @@ namespace detail ...@@ -339,7 +348,7 @@ namespace detail
condition_impl *prepare(const category &c) override condition_impl *prepare(const category &c) override
{ {
m_item_ix = get_column_ix(c, m_item_tag); m_item_ix = get_item_ix(c, m_item_tag);
return this; return this;
} }
...@@ -415,8 +424,8 @@ namespace detail ...@@ -415,8 +424,8 @@ namespace detail
condition_impl *prepare(const category &c) override condition_impl *prepare(const category &c) override
{ {
m_item_ix = get_column_ix(c, m_item_tag); m_item_ix = get_item_ix(c, m_item_tag);
m_icase = is_column_type_uchar(c, m_item_tag); m_icase = is_item_type_uchar(c, m_item_tag);
return this; return this;
} }
...@@ -473,8 +482,8 @@ namespace detail ...@@ -473,8 +482,8 @@ namespace detail
condition_impl *prepare(const category &c) override condition_impl *prepare(const category &c) override
{ {
m_item_ix = get_column_ix(c, m_item_tag); m_item_ix = get_item_ix(c, m_item_tag);
m_icase = is_column_type_uchar(c, m_item_tag); m_icase = is_item_type_uchar(c, m_item_tag);
return this; return this;
} }
...@@ -506,7 +515,7 @@ namespace detail ...@@ -506,7 +515,7 @@ namespace detail
condition_impl *prepare(const category &c) override condition_impl *prepare(const category &c) override
{ {
m_item_ix = get_column_ix(c, m_item_tag); m_item_ix = get_item_ix(c, m_item_tag);
return this; return this;
} }
...@@ -541,7 +550,7 @@ namespace detail ...@@ -541,7 +550,7 @@ namespace detail
auto &c = r.get_category(); auto &c = r.get_category();
bool result = false; bool result = false;
for (auto &f : get_category_fields(c)) for (auto &f : get_category_items(c))
{ {
try try
{ {
...@@ -579,7 +588,7 @@ namespace detail ...@@ -579,7 +588,7 @@ namespace detail
auto &c = r.get_category(); auto &c = r.get_category();
bool result = false; bool result = false;
for (auto &f : get_category_fields(c)) for (auto &f : get_category_items(c))
{ {
try try
{ {
...@@ -887,7 +896,7 @@ inline condition operator or(condition &&a, condition &&b) ...@@ -887,7 +896,7 @@ inline condition operator or(condition &&a, condition &&b)
} }
/** /**
* @brief A helper class to make it possible to search for empty fields (NULL) * @brief A helper class to make it possible to search for empty items (NULL)
* *
* @code{.cpp} * @code{.cpp}
* "id"_key == cif::empty_type(); * "id"_key == cif::empty_type();
...@@ -909,7 +918,7 @@ struct empty_type ...@@ -909,7 +918,7 @@ struct empty_type
inline constexpr empty_type null = empty_type(); inline constexpr empty_type null = empty_type();
/** /**
* @brief Class to use in creating conditions, creates a reference to a field or column * @brief Class to use in creating conditions, creates a reference to a item or item
* *
*/ */
struct key struct key
...@@ -947,7 +956,7 @@ struct key ...@@ -947,7 +956,7 @@ struct key
key(const key &) = delete; key(const key &) = delete;
key &operator=(const key &) = delete; key &operator=(const key &) = delete;
std::string m_item_tag; ///< The column name std::string m_item_tag; ///< The item name
}; };
/** /**
...@@ -1072,7 +1081,7 @@ inline condition operator!=(const key &key, const empty_type &) ...@@ -1072,7 +1081,7 @@ inline condition operator!=(const key &key, const empty_type &)
} }
/** /**
* @brief Create a condition to search any column for a value @a v if @a v contains a value * @brief Create a condition to search any item for a value @a v if @a v contains a value
* compare to null if not. * compare to null if not.
*/ */
template <typename T> template <typename T>
...@@ -1099,12 +1108,12 @@ struct any_type ...@@ -1099,12 +1108,12 @@ struct any_type
/** @endcond */ /** @endcond */
/** /**
* @brief A helper for any field constructs * @brief A helper for any item constructs
*/ */
inline constexpr any_type any = any_type{}; inline constexpr any_type any = any_type{};
/** /**
* @brief Create a condition to search any column for a value @a v * @brief Create a condition to search any item for a value @a v
*/ */
template <typename T> template <typename T>
condition operator==(const any_type &, const T &v) condition operator==(const any_type &, const T &v)
...@@ -1113,7 +1122,7 @@ condition operator==(const any_type &, const T &v) ...@@ -1113,7 +1122,7 @@ condition operator==(const any_type &, const T &v)
} }
/** /**
* @brief Create a condition to search any column for a regular expression @a rx * @brief Create a condition to search any item for a regular expression @a rx
*/ */
inline condition operator==(const any_type &, const std::regex &rx) inline condition operator==(const any_type &, const std::regex &rx)
{ {
...@@ -1131,9 +1140,9 @@ inline condition all() ...@@ -1131,9 +1140,9 @@ inline condition all()
namespace literals namespace literals
{ {
/** /**
* @brief Return a cif::key for the column name @a text * @brief Return a cif::key for the item name @a text
* *
* @param text The name of the column * @param text The name of the item
* @param length The length of @a text * @param length The length of @a text
* @return key The cif::key created * @return key The cif::key created
*/ */
......
include/cif++/item.hpp
...@@ -44,7 +44,7 @@ ...@@ -44,7 +44,7 @@
/** \file item.hpp /** \file item.hpp
* *
* This file contains the declaration of item but also the item_value and item_handle * This file contains the declaration of item but also the item_value and item_handle
* These handle the storage of and access to the data for a single data field. * These handle the storage of and access to the data for a single data item.
*/ */
namespace cif namespace cif
...@@ -227,10 +227,10 @@ class item ...@@ -227,10 +227,10 @@ class item
/// \brief empty means either null or unknown /// \brief empty means either null or unknown
bool empty() const { return m_value.empty(); } bool empty() const { return m_value.empty(); }
/// \brief returns true if the field contains '.' /// \brief returns true if the item contains '.'
bool is_null() const { return m_value == "."; } bool is_null() const { return m_value == "."; }
/// \brief returns true if the field contains '?' /// \brief returns true if the item contains '?'
bool is_unknown() const { return m_value == "?"; } bool is_unknown() const { return m_value == "?"; }
/// \brief the length of the value string /// \brief the length of the value string
...@@ -464,14 +464,14 @@ struct item_handle ...@@ -464,14 +464,14 @@ struct item_handle
/** Easy way to test for an empty item */ /** Easy way to test for an empty item */
explicit operator bool() const { return not empty(); } explicit operator bool() const { return not empty(); }
/// is_null return true if the field contains '.' /// is_null return true if the item contains '.'
bool is_null() const bool is_null() const
{ {
auto txt = text(); auto txt = text();
return txt.length() == 1 and txt.front() == '.'; return txt.length() == 1 and txt.front() == '.';
} }
/// is_unknown returns true if the field contains '?' /// is_unknown returns true if the item contains '?'
bool is_unknown() const bool is_unknown() const
{ {
auto txt = text(); auto txt = text();
...@@ -484,11 +484,11 @@ struct item_handle ...@@ -484,11 +484,11 @@ struct item_handle
/** /**
* @brief Construct a new item handle object * @brief Construct a new item handle object
* *
* @param column Column index * @param item Item index
* @param row Reference to the row * @param row Reference to the row
*/ */
item_handle(uint16_t column, row_handle &row) item_handle(uint16_t item, row_handle &row)
: m_column(column) : m_item_ix(item)
, m_row_handle(row) , m_row_handle(row)
{ {
} }
...@@ -505,7 +505,7 @@ struct item_handle ...@@ -505,7 +505,7 @@ struct item_handle
private: private:
item_handle(); item_handle();
uint16_t m_column; uint16_t m_item_ix;
row_handle &m_row_handle; row_handle &m_row_handle;
void assign_value(const item &value); void assign_value(const item &value);
......
include/cif++/iterator.hpp
...@@ -90,7 +90,7 @@ class iterator_impl ...@@ -90,7 +90,7 @@ class iterator_impl
: m_category(rhs.m_category) : m_category(rhs.m_category)
, m_current(rhs.m_current) , m_current(rhs.m_current)
, m_value(rhs.m_value) , m_value(rhs.m_value)
, m_column_ix(rhs.m_column_ix) , m_item_ix(rhs.m_item_ix)
{ {
} }
...@@ -99,7 +99,7 @@ class iterator_impl ...@@ -99,7 +99,7 @@ class iterator_impl
: m_category(rhs.m_category) : m_category(rhs.m_category)
, m_current(const_cast<row_type *>(rhs.m_current)) , m_current(const_cast<row_type *>(rhs.m_current))
, m_value(rhs.m_value) , m_value(rhs.m_value)
, m_column_ix(rhs.m_column_ix) , m_item_ix(rhs.m_item_ix)
{ {
m_value = get(std::make_index_sequence<N>()); m_value = get(std::make_index_sequence<N>());
} }
...@@ -108,7 +108,7 @@ class iterator_impl ...@@ -108,7 +108,7 @@ class iterator_impl
iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<uint16_t, N> &cix) iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<uint16_t, N> &cix)
: m_category(rhs.m_category) : m_category(rhs.m_category)
, m_current(rhs.m_current) , m_current(rhs.m_current)
, m_column_ix(cix) , m_item_ix(cix)
{ {
m_value = get(std::make_index_sequence<N>()); m_value = get(std::make_index_sequence<N>());
} }
...@@ -117,7 +117,7 @@ class iterator_impl ...@@ -117,7 +117,7 @@ class iterator_impl
{ {
m_category = i.m_category; m_category = i.m_category;
m_current = i.m_current; m_current = i.m_current;
m_column_ix = i.m_column_ix; m_item_ix = i.m_item_ix;
m_value = i.m_value; m_value = i.m_value;
return *this; return *this;
} }
...@@ -185,7 +185,7 @@ class iterator_impl ...@@ -185,7 +185,7 @@ class iterator_impl
if (m_current != nullptr) if (m_current != nullptr)
{ {
row_handle rh{ *m_category, *m_current }; row_handle rh{ *m_category, *m_current };
return tuple_type{ rh[m_column_ix[Is]].template as<Ts>()... }; return tuple_type{ rh[m_item_ix[Is]].template as<Ts>()... };
} }
return {}; return {};
...@@ -194,7 +194,7 @@ class iterator_impl ...@@ -194,7 +194,7 @@ class iterator_impl
category_type *m_category = nullptr; category_type *m_category = nullptr;
row_type *m_current = nullptr; row_type *m_current = nullptr;
value_type m_value; value_type m_value;
std::array<uint16_t, N> m_column_ix; std::array<uint16_t, N> m_item_ix;
}; };
/** /**
...@@ -348,7 +348,7 @@ class iterator_impl<Category, T> ...@@ -348,7 +348,7 @@ class iterator_impl<Category, T>
: m_category(rhs.m_category) : m_category(rhs.m_category)
, m_current(rhs.m_current) , m_current(rhs.m_current)
, m_value(rhs.m_value) , m_value(rhs.m_value)
, m_column_ix(rhs.m_column_ix) , m_item_ix(rhs.m_item_ix)
{ {
} }
...@@ -357,7 +357,7 @@ class iterator_impl<Category, T> ...@@ -357,7 +357,7 @@ class iterator_impl<Category, T>
: m_category(rhs.m_category) : m_category(rhs.m_category)
, m_current(const_cast<row_type *>(rhs.m_current)) , m_current(const_cast<row_type *>(rhs.m_current))
, m_value(rhs.m_value) , m_value(rhs.m_value)
, m_column_ix(rhs.m_column_ix) , m_item_ix(rhs.m_item_ix)
{ {
m_value = get(m_current); m_value = get(m_current);
} }
...@@ -366,7 +366,7 @@ class iterator_impl<Category, T> ...@@ -366,7 +366,7 @@ class iterator_impl<Category, T>
iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<uint16_t, 1> &cix) iterator_impl(const iterator_impl<IRowType> &rhs, const std::array<uint16_t, 1> &cix)
: m_category(rhs.m_category) : m_category(rhs.m_category)
, m_current(rhs.m_current) , m_current(rhs.m_current)
, m_column_ix(cix[0]) , m_item_ix(cix[0])
{ {
m_value = get(); m_value = get();
} }
...@@ -375,7 +375,7 @@ class iterator_impl<Category, T> ...@@ -375,7 +375,7 @@ class iterator_impl<Category, T>
{ {
m_category = i.m_category; m_category = i.m_category;
m_current = i.m_current; m_current = i.m_current;
m_column_ix = i.m_column_ix; m_item_ix = i.m_item_ix;
m_value = i.m_value; m_value = i.m_value;
return *this; return *this;
} }
...@@ -442,7 +442,7 @@ class iterator_impl<Category, T> ...@@ -442,7 +442,7 @@ class iterator_impl<Category, T>
if (m_current != nullptr) if (m_current != nullptr)
{ {
row_handle rh{ *m_category, *m_current }; row_handle rh{ *m_category, *m_current };
return rh[m_column_ix].template as<T>(); return rh[m_item_ix].template as<T>();
} }
return {}; return {};
...@@ -451,7 +451,7 @@ class iterator_impl<Category, T> ...@@ -451,7 +451,7 @@ class iterator_impl<Category, T>
category_type *m_category = nullptr; category_type *m_category = nullptr;
row_type *m_current = nullptr; row_type *m_current = nullptr;
value_type m_value; value_type m_value;
uint16_t m_column_ix; uint16_t m_item_ix;
}; };
// -------------------------------------------------------------------- // --------------------------------------------------------------------
...@@ -482,8 +482,8 @@ class iterator_proxy ...@@ -482,8 +482,8 @@ class iterator_proxy
using iterator = iterator_impl<category_type, Ts...>; using iterator = iterator_impl<category_type, Ts...>;
using row_iterator = iterator_impl<category_type>; using row_iterator = iterator_impl<category_type>;
iterator_proxy(category_type &cat, row_iterator pos, char const *const columns[N]); iterator_proxy(category_type &cat, row_iterator pos, char const *const items[N]);
iterator_proxy(category_type &cat, row_iterator pos, std::initializer_list<char const *> columns); iterator_proxy(category_type &cat, row_iterator pos, std::initializer_list<char const *> items);
iterator_proxy(iterator_proxy &&p); iterator_proxy(iterator_proxy &&p);
iterator_proxy &operator=(iterator_proxy &&p); iterator_proxy &operator=(iterator_proxy &&p);
...@@ -492,8 +492,8 @@ class iterator_proxy ...@@ -492,8 +492,8 @@ class iterator_proxy
iterator_proxy &operator=(const iterator_proxy &) = delete; iterator_proxy &operator=(const iterator_proxy &) = delete;
/** @endcond */ /** @endcond */
iterator begin() const { return iterator(m_begin, m_column_ix); } ///< Return the iterator pointing to the first row iterator begin() const { return iterator(m_begin, m_item_ix); } ///< Return the iterator pointing to the first row
iterator end() const { return iterator(m_end, m_column_ix); } ///< Return the iterator pointing past the last row iterator end() const { return iterator(m_end, m_item_ix); } ///< Return the iterator pointing past the last row
bool empty() const { return m_begin == m_end; } ///< Return true if the range is empty bool empty() const { return m_begin == m_end; } ///< Return true if the range is empty
explicit operator bool() const { return not empty(); } ///< Easy way to detect if the range is empty explicit operator bool() const { return not empty(); } ///< Easy way to detect if the range is empty
...@@ -510,13 +510,13 @@ class iterator_proxy ...@@ -510,13 +510,13 @@ class iterator_proxy
std::swap(m_category, rhs.m_category); std::swap(m_category, rhs.m_category);
std::swap(m_begin, rhs.m_begin); std::swap(m_begin, rhs.m_begin);
std::swap(m_end, rhs.m_end); std::swap(m_end, rhs.m_end);
std::swap(m_column_ix, rhs.m_column_ix); std::swap(m_item_ix, rhs.m_item_ix);
} }
private: private:
category_type *m_category; category_type *m_category;
row_iterator m_begin, m_end; row_iterator m_begin, m_end;
std::array<uint16_t, N> m_column_ix; std::array<uint16_t, N> m_item_ix;
}; };
// -------------------------------------------------------------------- // --------------------------------------------------------------------
...@@ -651,26 +651,26 @@ class conditional_iterator_proxy ...@@ -651,26 +651,26 @@ class conditional_iterator_proxy
/** @cond */ /** @cond */
template <typename Category, typename... Ts> template <typename Category, typename... Ts>
iterator_proxy<Category, Ts...>::iterator_proxy(Category &cat, row_iterator pos, char const *const columns[N]) iterator_proxy<Category, Ts...>::iterator_proxy(Category &cat, row_iterator pos, char const *const items[N])
: m_category(&cat) : m_category(&cat)
, m_begin(pos) , m_begin(pos)
, m_end(cat.end()) , m_end(cat.end())
{ {
for (uint16_t i = 0; i < N; ++i) for (uint16_t i = 0; i < N; ++i)
m_column_ix[i] = m_category->get_column_ix(columns[i]); m_item_ix[i] = m_category->get_item_ix(items[i]);
} }
template <typename Category, typename... Ts> template <typename Category, typename... Ts>
iterator_proxy<Category, Ts...>::iterator_proxy(Category &cat, row_iterator pos, std::initializer_list<char const *> columns) iterator_proxy<Category, Ts...>::iterator_proxy(Category &cat, row_iterator pos, std::initializer_list<char const *> items)
: m_category(&cat) : m_category(&cat)
, m_begin(pos) , m_begin(pos)
, m_end(cat.end()) , m_end(cat.end())
{ {
// static_assert(columns.size() == N, "The list of column names should be exactly the same as the list of requested columns"); // static_assert(items.size() == N, "The list of item names should be exactly the same as the list of requested items");
std::uint16_t i = 0; std::uint16_t i = 0;
for (auto column : columns) for (auto item : items)
m_column_ix[i++] = m_category->get_column_ix(column); m_item_ix[i++] = m_category->get_item_ix(item);
} }
// -------------------------------------------------------------------- // --------------------------------------------------------------------
...@@ -707,7 +707,7 @@ conditional_iterator_proxy<Category, Ts...>::conditional_iterator_proxy(Category ...@@ -707,7 +707,7 @@ conditional_iterator_proxy<Category, Ts...>::conditional_iterator_proxy(Category
, mCBegin(pos) , mCBegin(pos)
, mCEnd(cat.end()) , mCEnd(cat.end())
{ {
static_assert(sizeof...(Ts) == sizeof...(Ns), "Number of column names should be equal to number of requested value types"); static_assert(sizeof...(Ts) == sizeof...(Ns), "Number of item names should be equal to number of requested value types");
if (m_condition) if (m_condition)
{ {
...@@ -720,7 +720,7 @@ conditional_iterator_proxy<Category, Ts...>::conditional_iterator_proxy(Category ...@@ -720,7 +720,7 @@ conditional_iterator_proxy<Category, Ts...>::conditional_iterator_proxy(Category
mCBegin = mCEnd; mCBegin = mCEnd;
uint16_t i = 0; uint16_t i = 0;
((mCix[i++] = m_cat->get_column_ix(names)), ...); ((mCix[i++] = m_cat->get_item_ix(names)), ...);
} }
template <typename Category, typename... Ts> template <typename Category, typename... Ts>
......
include/cif++/model.hpp
...@@ -72,7 +72,7 @@ class structure; ...@@ -72,7 +72,7 @@ class structure;
* *
* The class atom is a kind of flyweight class. It can be copied * The class atom is a kind of flyweight class. It can be copied
* with low overhead. All data is stored in the underlying mmCIF * with low overhead. All data is stored in the underlying mmCIF
* categories but some very often used fields are cached in the * categories but some very often used items are cached in the
* impl. * impl.
* *
* It is also possible to have symmetry copies of atoms. They * It is also possible to have symmetry copies of atoms. They
...@@ -207,7 +207,7 @@ class atom ...@@ -207,7 +207,7 @@ class atom
/// \brief Copy assignement operator /// \brief Copy assignement operator
atom &operator=(const atom &rhs) = default; atom &operator=(const atom &rhs) = default;
/// \brief Return the field named @a name in the _atom_site category for this atom /// \brief Return the item named @a name in the _atom_site category for this atom
std::string get_property(std::string_view name) const std::string get_property(std::string_view name) const
{ {
if (not m_impl) if (not m_impl)
...@@ -215,7 +215,7 @@ class atom ...@@ -215,7 +215,7 @@ class atom
return m_impl->get_property(name); return m_impl->get_property(name);
} }
/// \brief Return the field named @a name in the _atom_site category for this atom cast to an int /// \brief Return the item named @a name in the _atom_site category for this atom cast to an int
int get_property_int(std::string_view name) const int get_property_int(std::string_view name) const
{ {
if (not m_impl) if (not m_impl)
...@@ -223,7 +223,7 @@ class atom ...@@ -223,7 +223,7 @@ class atom
return m_impl->get_property_int(name); return m_impl->get_property_int(name);
} }
/// \brief Return the field named @a name in the _atom_site category for this atom cast to a float /// \brief Return the item named @a name in the _atom_site category for this atom cast to a float
float get_property_float(std::string_view name) const float get_property_float(std::string_view name) const
{ {
if (not m_impl) if (not m_impl)
...@@ -231,7 +231,7 @@ class atom ...@@ -231,7 +231,7 @@ class atom
return m_impl->get_property_float(name); return m_impl->get_property_float(name);
} }
/// \brief Set value for the field named @a name in the _atom_site category to @a value /// \brief Set value for the item named @a name in the _atom_site category to @a value
void set_property(const std::string_view name, const std::string &value) void set_property(const std::string_view name, const std::string &value)
{ {
if (not m_impl) if (not m_impl)
...@@ -239,7 +239,7 @@ class atom ...@@ -239,7 +239,7 @@ class atom
m_impl->set_property(name, value); m_impl->set_property(name, value);
} }
/// \brief Set value for the field named @a name in the _atom_site category to @a value /// \brief Set value for the item named @a name in the _atom_site category to @a value
template <typename T, std::enable_if_t<std::is_arithmetic_v<T>, int> = 0> template <typename T, std::enable_if_t<std::is_arithmetic_v<T>, int> = 0>
void set_property(const std::string_view name, const T &value) void set_property(const std::string_view name, const T &value)
{ {
...@@ -730,7 +730,7 @@ class sugar : public residue ...@@ -730,7 +730,7 @@ class sugar : public residue
/** /**
* @brief Return the sugar number in the glycosylation tree * @brief Return the sugar number in the glycosylation tree
* *
* To store the sugar number, the auth_seq_id field has been overloaded * To store the sugar number, the auth_seq_id item has been overloaded
* in the specification. But since a sugar number should be, ehm, a number * in the specification. But since a sugar number should be, ehm, a number
* and auth_seq_id is specified to contain a string, we do a check here * and auth_seq_id is specified to contain a string, we do a check here
* to see if it really is a number. * to see if it really is a number.
......
include/cif++/parser.hpp
...@@ -268,8 +268,8 @@ class sac_parser ...@@ -268,8 +268,8 @@ class sac_parser
QuotedStringQuote, QuotedStringQuote,
UnquotedString, UnquotedString,
Tag, Tag,
TextField, TextItem,
TextFieldNL, TextItemNL,
Reserved, Reserved,
Value Value
}; };
......
include/cif++/pdb.hpp
...@@ -28,6 +28,8 @@ ...@@ -28,6 +28,8 @@
#include "cif++/file.hpp" #include "cif++/file.hpp"
#include <system_error>
/** /**
* @file pdb.hpp * @file pdb.hpp
* *
...@@ -120,6 +122,8 @@ void reconstruct_pdbx(file &pdbx_file, std::string_view dictionary = "mmcif_pdbx ...@@ -120,6 +122,8 @@ void reconstruct_pdbx(file &pdbx_file, std::string_view dictionary = "mmcif_pdbx
* *
* Use the common \ref cif::VERBOSE flag to turn on diagnostic messages. * Use the common \ref cif::VERBOSE flag to turn on diagnostic messages.
* *
* This function throws a std::system_error in case of an error
*
* \param file The input file * \param file The input file
* \param dictionary The mmcif dictionary to use * \param dictionary The mmcif dictionary to use
* \result Returns true if the file was valid and consistent * \result Returns true if the file was valid and consistent
...@@ -127,6 +131,43 @@ void reconstruct_pdbx(file &pdbx_file, std::string_view dictionary = "mmcif_pdbx ...@@ -127,6 +131,43 @@ void reconstruct_pdbx(file &pdbx_file, std::string_view dictionary = "mmcif_pdbx
bool is_valid_pdbx_file(const file &pdbx_file, std::string_view dictionary = "mmcif_pdbx"); bool is_valid_pdbx_file(const file &pdbx_file, std::string_view dictionary = "mmcif_pdbx");
/** \brief This is an extension to cif::validator, use the logic in common
* PDBx files to see if the file is internally consistent.
*
* This function for now checks if the following categories are consistent:
*
* atom_site -> pdbx_poly_seq_scheme -> entity_poly_seq -> entity_poly -> entity
*
* Use the common \ref cif::VERBOSE flag to turn on diagnostic messages.
*
* The dictionary is assumed to be specified in the file or to be the
* default mmcif_pdbx.dic dictionary.
*
* \param file The input file
* \param ec The error_code in case something was wrong
* \result Returns true if the file was valid and consistent
*/
bool is_valid_pdbx_file(const file &pdbx_file, std::error_code &ec);
/** \brief This is an extension to cif::validator, use the logic in common
* PDBx files to see if the file is internally consistent.
*
* This function for now checks if the following categories are consistent:
*
* atom_site -> pdbx_poly_seq_scheme -> entity_poly_seq -> entity_poly -> entity
*
* Use the common \ref cif::VERBOSE flag to turn on diagnostic messages.
*
* \param file The input file
* \param dictionary The dictionary to use
* \param ec The error_code in case something was wrong
* \result Returns true if the file was valid and consistent
*/
bool is_valid_pdbx_file(const file &pdbx_file, std::string_view dictionary,
std::error_code &ec);
// -------------------------------------------------------------------- // --------------------------------------------------------------------
// Other I/O related routines // Other I/O related routines
......
include/cif++/row.hpp
...@@ -51,7 +51,7 @@ ...@@ -51,7 +51,7 @@
* std::string name = rh["label_atom_id"].as<std::string>(); * std::string name = rh["label_atom_id"].as<std::string>();
* *
* // by index: * // by index:
* uint16_t ix = atom_site.get_column_ix("label_atom_id"); * uint16_t ix = atom_site.get_item_ix("label_atom_id");
* assert(rh[ix].as<std::string() == name); * assert(rh[ix].as<std::string() == name);
* @endcode * @endcode
* *
...@@ -87,15 +87,15 @@ namespace detail ...@@ -87,15 +87,15 @@ namespace detail
{ {
static constexpr size_t N = sizeof...(C); static constexpr size_t N = sizeof...(C);
get_row_result(const row_handle &r, std::array<uint16_t, N> &&columns) get_row_result(const row_handle &r, std::array<uint16_t, N> &&items)
: m_row(r) : m_row(r)
, m_columns(std::move(columns)) , m_items(std::move(items))
{ {
} }
const item_handle operator[](uint16_t ix) const const item_handle operator[](uint16_t ix) const
{ {
return m_row[m_columns[ix]]; return m_row[m_items[ix]];
} }
template <typename... Ts, std::enable_if_t<N == sizeof...(Ts), int> = 0> template <typename... Ts, std::enable_if_t<N == sizeof...(Ts), int> = 0>
...@@ -107,11 +107,11 @@ namespace detail ...@@ -107,11 +107,11 @@ namespace detail
template <typename... Ts, size_t... Is> template <typename... Ts, size_t... Is>
std::tuple<Ts...> get(std::index_sequence<Is...>) const std::tuple<Ts...> get(std::index_sequence<Is...>) const
{ {
return std::tuple<Ts...>{ m_row[m_columns[Is]].template as<Ts>()... }; return std::tuple<Ts...>{ m_row[m_items[Is]].template as<Ts>()... };
} }
const row_handle &m_row; const row_handle &m_row;
std::array<uint16_t, N> m_columns; std::array<uint16_t, N> m_items;
}; };
// we want to be able to tie some variables to a get_row_result, for this we use tiewraps // we want to be able to tie some variables to a get_row_result, for this we use tiewraps
...@@ -244,70 +244,70 @@ class row_handle ...@@ -244,70 +244,70 @@ class row_handle
return not empty(); return not empty();
} }
/// \brief return a cif::item_handle to the item in column @a column_ix /// \brief return a cif::item_handle to the item in item @a item_ix
item_handle operator[](uint16_t column_ix) item_handle operator[](uint16_t item_ix)
{ {
return empty() ? item_handle::s_null_item : item_handle(column_ix, *this); return empty() ? item_handle::s_null_item : item_handle(item_ix, *this);
} }
/// \brief return a const cif::item_handle to the item in column @a column_ix /// \brief return a const cif::item_handle to the item in item @a item_ix
const item_handle operator[](uint16_t column_ix) const const item_handle operator[](uint16_t item_ix) const
{ {
return empty() ? item_handle::s_null_item : item_handle(column_ix, const_cast<row_handle &>(*this)); return empty() ? item_handle::s_null_item : item_handle(item_ix, const_cast<row_handle &>(*this));
} }
/// \brief return a cif::item_handle to the item in the column named @a column_name /// \brief return a cif::item_handle to the item in the item named @a item_name
item_handle operator[](std::string_view column_name) item_handle operator[](std::string_view item_name)
{ {
return empty() ? item_handle::s_null_item : item_handle(add_column(column_name), *this); return empty() ? item_handle::s_null_item : item_handle(add_item(item_name), *this);
} }
/// \brief return a const cif::item_handle to the item in the column named @a column_name /// \brief return a const cif::item_handle to the item in the item named @a item_name
const item_handle operator[](std::string_view column_name) const const item_handle operator[](std::string_view item_name) const
{ {
return empty() ? item_handle::s_null_item : item_handle(get_column_ix(column_name), const_cast<row_handle &>(*this)); return empty() ? item_handle::s_null_item : item_handle(get_item_ix(item_name), const_cast<row_handle &>(*this));
} }
/// \brief Return an object that can be used in combination with cif::tie /// \brief Return an object that can be used in combination with cif::tie
/// to assign the values for the columns @a columns /// to assign the values for the items @a items
template <typename... C> template <typename... C>
auto get(C... columns) const auto get(C... items) const
{ {
return detail::get_row_result<C...>(*this, { get_column_ix(columns)... }); return detail::get_row_result<C...>(*this, { get_item_ix(items)... });
} }
/// \brief Return a tuple of values of types @a Ts for the columns @a columns /// \brief Return a tuple of values of types @a Ts for the items @a items
template <typename... Ts, typename... C, std::enable_if_t<sizeof...(Ts) == sizeof...(C) and sizeof...(C) != 1, int> = 0> template <typename... Ts, typename... C, std::enable_if_t<sizeof...(Ts) == sizeof...(C) and sizeof...(C) != 1, int> = 0>
std::tuple<Ts...> get(C... columns) const std::tuple<Ts...> get(C... items) const
{ {
return detail::get_row_result<Ts...>(*this, { get_column_ix(columns)... }); return detail::get_row_result<Ts...>(*this, { get_item_ix(items)... });
} }
/// \brief Get the value of column @a column cast to type @a T /// \brief Get the value of item @a item cast to type @a T
template <typename T> template <typename T>
T get(const char *column) const T get(const char *item) const
{ {
return operator[](get_column_ix(column)).template as<T>(); return operator[](get_item_ix(item)).template as<T>();
} }
/// \brief Get the value of column @a column cast to type @a T /// \brief Get the value of item @a item cast to type @a T
template <typename T> template <typename T>
T get(std::string_view column) const T get(std::string_view item) const
{ {
return operator[](get_column_ix(column)).template as<T>(); return operator[](get_item_ix(item)).template as<T>();
} }
/// \brief assign each of the columns named in @a values to their respective value /// \brief assign each of the items named in @a values to their respective value
void assign(const std::vector<item> &values) void assign(const std::vector<item> &values)
{ {
for (auto &value : values) for (auto &value : values)
assign(value, true); assign(value, true);
} }
/** \brief assign the value @a value to the column named @a name /** \brief assign the value @a value to the item named @a name
* *
* If updateLinked it true, linked records are updated as well. * If updateLinked it true, linked records are updated as well.
* That means that if column @a name is part of the link definition * That means that if item @a name is part of the link definition
* and the link results in a linked record in another category * and the link results in a linked record in another category
* this record in the linked category is updated as well. * this record in the linked category is updated as well.
* *
...@@ -317,13 +317,13 @@ class row_handle ...@@ -317,13 +317,13 @@ class row_handle
void assign(std::string_view name, std::string_view value, bool updateLinked, bool validate = true) void assign(std::string_view name, std::string_view value, bool updateLinked, bool validate = true)
{ {
assign(add_column(name), value, updateLinked, validate); assign(add_item(name), value, updateLinked, validate);
} }
/** \brief assign the value @a value to column at index @a column /** \brief assign the value @a value to item at index @a item
* *
* If updateLinked it true, linked records are updated as well. * If updateLinked it true, linked records are updated as well.
* That means that if column @a column is part of the link definition * That means that if item @a item is part of the link definition
* and the link results in a linked record in another category * and the link results in a linked record in another category
* this record in the linked category is updated as well. * this record in the linked category is updated as well.
* *
...@@ -331,7 +331,7 @@ class row_handle ...@@ -331,7 +331,7 @@ class row_handle
* checked to see if it conforms to the rules defined in the dictionary * checked to see if it conforms to the rules defined in the dictionary
*/ */
void assign(uint16_t column, std::string_view value, bool updateLinked, bool validate = true); void assign(uint16_t item, std::string_view value, bool updateLinked, bool validate = true);
/// \brief compare two rows /// \brief compare two rows
bool operator==(const row_handle &rhs) const { return m_category == rhs.m_category and m_row == rhs.m_row; } bool operator==(const row_handle &rhs) const { return m_category == rhs.m_category and m_row == rhs.m_row; }
...@@ -340,10 +340,10 @@ class row_handle ...@@ -340,10 +340,10 @@ class row_handle
bool operator!=(const row_handle &rhs) const { return m_category != rhs.m_category or m_row != rhs.m_row; } bool operator!=(const row_handle &rhs) const { return m_category != rhs.m_category or m_row != rhs.m_row; }
private: private:
uint16_t get_column_ix(std::string_view name) const; uint16_t get_item_ix(std::string_view name) const;
std::string_view get_column_name(uint16_t ix) const; std::string_view get_item_name(uint16_t ix) const;
uint16_t add_column(std::string_view name); uint16_t add_item(std::string_view name);
row *get_row() row *get_row()
{ {
...@@ -360,7 +360,7 @@ class row_handle ...@@ -360,7 +360,7 @@ class row_handle
assign(i.name(), i.value(), updateLinked); assign(i.name(), i.value(), updateLinked);
} }
void swap(uint16_t column, row_handle &r); void swap(uint16_t item, row_handle &r);
category *m_category = nullptr; category *m_category = nullptr;
row *m_row = nullptr; row *m_row = nullptr;
......
include/cif++/validate.hpp
...@@ -28,9 +28,11 @@ ...@@ -28,9 +28,11 @@
#include "cif++/text.hpp" #include "cif++/text.hpp"
#include <cassert>
#include <filesystem> #include <filesystem>
#include <list> #include <list>
#include <mutex> #include <mutex>
#include <system_error>
#include <utility> #include <utility>
/** /**
...@@ -49,29 +51,123 @@ namespace cif ...@@ -49,29 +51,123 @@ namespace cif
struct category_validator; struct category_validator;
// -------------------------------------------------------------------- // --------------------------------------------------------------------
// New: error_code
/** /**
* @brief The exception thrown when a validation error occurs * @enum validation_error
* *
* @brief A stronly typed class containing the error codes reported by @ref cif::validator and friends
*/ */
class validation_error : public std::exception enum class validation_error
{
value_does_not_match_rx = 1, /**< The value of an item does not conform to the regular expression specified for it */
value_is_not_in_enumeration_list, /**< The value of an item is not in the list of values allowed */
not_a_known_primitive_type, /**< The type is not a known primitive type */
undefined_category, /**< Category has no definition in the dictionary */
unknown_item, /**< The item is not defined to be part of the category */
incorrect_item_validator, /**< Incorrectly specified validator for item */
missing_mandatory_items, /**< Missing mandatory items */
missing_key_items, /**< An index could not be constructed due to missing key items */
item_not_allowed_in_category, /**< Requested item allowed in category according to dictionary */
empty_file, /**< The file contains no datablocks */
empty_datablock, /**< The datablock contains no categories */
empty_category, /**< The category is empty */
not_valid_pdbx, /**< The file is not a valid PDBx file */
};
/**
* @brief The implementation for @ref validation_category error messages
*
*/
class validation_category_impl : public std::error_category
{ {
public: public:
/// @brief Constructor /**
validation_error(const std::string &msg); * @brief User friendly name
*
* @return const char*
*/
/// @brief Constructor const char *name() const noexcept override
validation_error(const std::string &cat, const std::string &item, {
const std::string &msg); return "cif::validation";
}
/// @brief The description of the error /**
const char *what() const noexcept { return m_msg.c_str(); } * @brief Provide the error message as a string for the error code @a ev
*
* @param ev The error code
* @return std::string
*/
std::string message(int ev) const override
{
switch (static_cast<validation_error>(ev))
{
case validation_error::value_does_not_match_rx:
return "Value in item does not match regular expression";
case validation_error::value_is_not_in_enumeration_list:
return "Value is not in the enumerated list of valid values";
case validation_error::not_a_known_primitive_type:
return "The type is not a known primitive type";
case validation_error::undefined_category:
return "Category has no definition in the dictionary";
case validation_error::unknown_item:
return "The item is not defined to be part of the category";
case validation_error::incorrect_item_validator:
return "Incorrectly specified validator for item";
case validation_error::missing_mandatory_items:
return "Missing mandatory items";
case validation_error::missing_key_items:
return "An index could not be constructed due to missing key items";
case validation_error::item_not_allowed_in_category:
return "Requested item allowed in category according to dictionary";
case validation_error::empty_file:
return "The file contains no datablocks";
case validation_error::empty_datablock:
return "The datablock contains no categories";
case validation_error::empty_category:
return "The category is empty";
case validation_error::not_valid_pdbx:
return "The file is not a valid PDBx file";
default:
assert(false);
return "unknown error code";
}
}
/**
* @brief Return whether two error codes are equivalent, always false in this case
*
*/
/// @cond bool equivalent(const std::error_code & /*code*/, int /*condition*/) const noexcept override
std::string m_msg; {
/// @endcond return false;
}
}; };
/**
* @brief Return the implementation for the validation_category
*
* @return std::error_category&
*/
inline std::error_category &validation_category()
{
static validation_category_impl instance;
return instance;
}
inline std::error_code make_error_code(validation_error e)
{
return std::error_code(static_cast<int>(e), validation_category());
}
inline std::error_condition make_error_condition(validation_error e)
{
return std::error_condition(static_cast<int>(e), validation_category());
}
// -------------------------------------------------------------------- // --------------------------------------------------------------------
/** @brief the primitive types known */ /** @brief the primitive types known */
...@@ -85,6 +181,9 @@ enum class DDL_PrimitiveType ...@@ -85,6 +181,9 @@ enum class DDL_PrimitiveType
/// @brief Return the DDL_PrimitiveType encoded in @a s /// @brief Return the DDL_PrimitiveType encoded in @a s
DDL_PrimitiveType map_to_primitive_type(std::string_view s); DDL_PrimitiveType map_to_primitive_type(std::string_view s);
/// @brief Return the DDL_PrimitiveType encoded in @a s, error reporting variant
DDL_PrimitiveType map_to_primitive_type(std::string_view s, std::error_code &ec) noexcept;
struct regex_impl; struct regex_impl;
/** /**
...@@ -188,8 +287,11 @@ struct item_validator ...@@ -188,8 +287,11 @@ struct item_validator
} }
/// @brief Validate the value in @a value for this item /// @brief Validate the value in @a value for this item
/// Will throw a validation_error exception if it fails /// Will throw a std::system_error exception if it fails
void operator()(std::string_view value) const; void operator()(std::string_view value) const;
/// @brief A more gentle version of value validation
bool validate_value(std::string_view value, std::error_code &ec) const noexcept;
}; };
/** /**
...@@ -203,7 +305,7 @@ struct category_validator ...@@ -203,7 +305,7 @@ struct category_validator
std::string m_name; ///< The name of the category std::string m_name; ///< The name of the category
std::vector<std::string> m_keys; ///< The list of items that make up the key std::vector<std::string> m_keys; ///< The list of items that make up the key
cif::iset m_groups; ///< The category groups this category belongs to cif::iset m_groups; ///< The category groups this category belongs to
cif::iset m_mandatory_fields; ///< The mandatory fields for this category cif::iset m_mandatory_items; ///< The mandatory items for this category
std::set<item_validator> m_item_validators; ///< The item validators for the items in this category std::set<item_validator> m_item_validators; ///< The item validators for the items in this category
/// @brief return true if this category sorts before @a rhs /// @brief return true if this category sorts before @a rhs
...@@ -298,7 +400,24 @@ class validator ...@@ -298,7 +400,24 @@ class validator
std::vector<const link_validator *> get_links_for_child(std::string_view category) const; std::vector<const link_validator *> get_links_for_child(std::string_view category) const;
/// @brief Bottleneck function to report an error in validation /// @brief Bottleneck function to report an error in validation
void report_error(const std::string &msg, bool fatal) const; void report_error(validation_error err, bool fatal = true) const
{
report_error(make_error_code(err), fatal);
}
/// @brief Bottleneck function to report an error in validation
void report_error(std::error_code ec, bool fatal = true) const;
/// @brief Bottleneck function to report an error in validation
void report_error(validation_error err, std::string_view category,
std::string_view item, bool fatal = true) const
{
report_error(make_error_code(err), category, item, fatal);
}
/// @brief Bottleneck function to report an error in validation
void report_error(std::error_code ec, std::string_view category,
std::string_view item, bool fatal = true) const;
const std::string &name() const { return m_name; } ///< Get the name of this validator const std::string &name() const { return m_name; } ///< Get the name of this validator
void set_name(const std::string &name) { m_name = name; } ///< Set the name of this validator void set_name(const std::string &name) { m_name = name; } ///< Set the name of this validator
......
src/condition.cpp
...@@ -30,17 +30,17 @@ ...@@ -30,17 +30,17 @@
namespace cif namespace cif
{ {
iset get_category_fields(const category &cat) iset get_category_items(const category &cat)
{ {
return cat.key_fields(); return cat.key_items();
} }
uint16_t get_column_ix(const category &cat, std::string_view col) uint16_t get_item_ix(const category &cat, std::string_view col)
{ {
return cat.get_column_ix(col); return cat.get_item_ix(col);
} }
bool is_column_type_uchar(const category &cat, std::string_view col) bool is_item_type_uchar(const category &cat, std::string_view col)
{ {
bool result = false; bool result = false;
...@@ -63,12 +63,12 @@ namespace detail ...@@ -63,12 +63,12 @@ namespace detail
condition_impl *key_equals_condition_impl::prepare(const category &c) condition_impl *key_equals_condition_impl::prepare(const category &c)
{ {
m_item_ix = c.get_column_ix(m_item_tag); m_item_ix = c.get_item_ix(m_item_tag);
m_icase = is_column_type_uchar(c, m_item_tag); m_icase = is_item_type_uchar(c, m_item_tag);
if (c.get_cat_validator() != nullptr and if (c.get_cat_validator() != nullptr and
c.key_field_indices().contains(m_item_ix) and c.key_item_indices().contains(m_item_ix) and
c.key_field_indices().size() == 1) c.key_item_indices().size() == 1)
{ {
m_single_hit = c[{ { m_item_tag, m_value } }]; m_single_hit = c[{ { m_item_tag, m_value } }];
} }
......
src/item.cpp
...@@ -35,7 +35,7 @@ const item_handle item_handle::s_null_item; ...@@ -35,7 +35,7 @@ const item_handle item_handle::s_null_item;
row_handle s_null_row_handle; row_handle s_null_row_handle;
item_handle::item_handle() item_handle::item_handle()
: m_column(std::numeric_limits<uint16_t>::max()) : m_item_ix(std::numeric_limits<uint16_t>::max())
, m_row_handle(s_null_row_handle) , m_row_handle(s_null_row_handle)
{ {
} }
...@@ -44,7 +44,7 @@ std::string_view item_handle::text() const ...@@ -44,7 +44,7 @@ std::string_view item_handle::text() const
{ {
if (not m_row_handle.empty()) if (not m_row_handle.empty())
{ {
auto iv = m_row_handle.m_row->get(m_column); auto iv = m_row_handle.m_row->get(m_item_ix);
if (iv != nullptr) if (iv != nullptr)
return iv->text(); return iv->text();
} }
...@@ -55,14 +55,14 @@ std::string_view item_handle::text() const ...@@ -55,14 +55,14 @@ std::string_view item_handle::text() const
void item_handle::assign_value(const item &v) void item_handle::assign_value(const item &v)
{ {
assert(not m_row_handle.empty()); assert(not m_row_handle.empty());
m_row_handle.assign(m_column, v.value(), true); m_row_handle.assign(m_item_ix, v.value(), true);
} }
void item_handle::swap(item_handle &b) void item_handle::swap(item_handle &b)
{ {
assert(m_column == b.m_column); assert(m_item_ix == b.m_item_ix);
// assert(&m_row_handle.m_category == &b.m_row_handle.m_category); // assert(&m_row_handle.m_category == &b.m_row_handle.m_category);
m_row_handle.swap(m_column, b.m_row_handle); m_row_handle.swap(m_item_ix, b.m_row_handle);
} }
} }
src/parser.cpp
...@@ -300,7 +300,7 @@ sac_parser::CIFToken sac_parser::get_next_token() ...@@ -300,7 +300,7 @@ sac_parser::CIFToken sac_parser::get_next_token()
else if (ch == '_') else if (ch == '_')
state = State::Tag; state = State::Tag;
else if (ch == ';' and m_bol) else if (ch == ';' and m_bol)
state = State::TextField; state = State::TextItem;
else if (ch == '?') else if (ch == '?')
state = State::QuestionMark; state = State::QuestionMark;
else if (ch == '\'' or ch == '"') else if (ch == '\'' or ch == '"')
...@@ -350,18 +350,18 @@ sac_parser::CIFToken sac_parser::get_next_token() ...@@ -350,18 +350,18 @@ sac_parser::CIFToken sac_parser::get_next_token()
state = State::Value; state = State::Value;
break; break;
case State::TextField: case State::TextItem:
if (ch == '\n') if (ch == '\n')
state = State::TextFieldNL; state = State::TextItemNL;
else if (ch == kEOF) else if (ch == kEOF)
error("unterminated textfield"); error("unterminated textfield");
else if (not is_any_print(ch) and cif::VERBOSE > 2) else if (not is_any_print(ch) and cif::VERBOSE > 2)
warning("invalid character in text field '" + std::string({static_cast<char>(ch)}) + "' (" + std::to_string((int)ch) + ")"); warning("invalid character in text field '" + std::string({static_cast<char>(ch)}) + "' (" + std::to_string((int)ch) + ")");
break; break;
case State::TextFieldNL: case State::TextItemNL:
if (is_text_lead(ch) or ch == ' ' or ch == '\t') if (is_text_lead(ch) or ch == ' ' or ch == '\t')
state = State::TextField; state = State::TextItem;
else if (ch == ';') else if (ch == ';')
{ {
assert(m_token_buffer.size() >= 2); assert(m_token_buffer.size() >= 2);
......
src/pdb/pdb2cif_remark_3.cpp
...@@ -1493,8 +1493,8 @@ bool Remark3Parser::parse(const std::string &expMethod, PDBRecord *r, cif::datab ...@@ -1493,8 +1493,8 @@ bool Remark3Parser::parse(const std::string &expMethod, PDBRecord *r, cif::datab
auto r1 = cat1.front(); auto r1 = cat1.front();
auto r2 = cat2.front(); auto r2 = cat2.front();
for (auto column : cat1.key_fields()) for (auto item : cat1.key_items())
r2[column] = r1[column].text(); r2[item] = r1[item].text();
} }
} }
else else
......
src/pdb/reconstruct.cpp
...@@ -237,7 +237,7 @@ void checkAtomRecords(datablock &db) ...@@ -237,7 +237,7 @@ void checkAtomRecords(datablock &db)
{ "auth_comp_id", auth_comp_id.value_or(*label_comp_id) }, { "auth_comp_id", auth_comp_id.value_or(*label_comp_id) },
{ "auth_atom_id", auth_atom_id.value_or(*label_atom_id) } }); { "auth_atom_id", auth_atom_id.value_or(*label_atom_id) } });
// Rewrite the coordinates and other fields that look better in a fixed format // Rewrite the coordinates and other items that look better in a fixed format
// Be careful not to nuke invalidly formatted data here // Be careful not to nuke invalidly formatted data here
for (auto [tag, prec] : std::vector<std::tuple<std::string_view, std::string::size_type>>{ for (auto [tag, prec] : std::vector<std::tuple<std::string_view, std::string::size_type>>{
{ "cartn_x", 3 }, { "cartn_x", 3 },
...@@ -267,8 +267,8 @@ void checkAtomRecords(datablock &db) ...@@ -267,8 +267,8 @@ void checkAtomRecords(datablock &db)
auto *cv = atom_site.get_cat_validator(); auto *cv = atom_site.get_cat_validator();
if (cv) if (cv)
{ {
// See if there are columns that are no longer known // See if there are items that are no longer known
for (auto tag : atom_site.get_columns()) for (auto tag : atom_site.get_items())
{ {
if (cv->get_validator_for_item(tag) != nullptr) if (cv->get_validator_for_item(tag) != nullptr)
continue; continue;
...@@ -277,12 +277,12 @@ void checkAtomRecords(datablock &db) ...@@ -277,12 +277,12 @@ void checkAtomRecords(datablock &db)
if (not r) if (not r)
{ {
if (cif::VERBOSE > 0) if (cif::VERBOSE > 0)
std::clog << "Dropping unknown column " << tag << '\n'; std::clog << "Dropping unknown item " << tag << '\n';
atom_site.remove_column(tag); atom_site.remove_item(tag);
} }
else if (cif::VERBOSE > 0) else if (cif::VERBOSE > 0)
std::clog << "Keeping unknown column " << std::quoted(tag) << " in atom_site since it is not empty\n"; std::clog << "Keeping unknown item " << std::quoted(tag) << " in atom_site since it is not empty\n";
} }
} }
} }
...@@ -311,10 +311,10 @@ void createEntity(datablock &db) ...@@ -311,10 +311,10 @@ void createEntity(datablock &db)
auto &cf = compound_factory::instance(); auto &cf = compound_factory::instance();
auto &atom_site = db["atom_site"]; auto &atom_site = db["atom_site"];
atom_site.add_column("label_entity_id"); atom_site.add_item("label_entity_id");
auto &struct_asym = db["struct_asym"]; auto &struct_asym = db["struct_asym"];
struct_asym.add_column("entity_id"); struct_asym.add_item("entity_id");
std::map<std::string, std::vector<std::tuple<std::string, int>>> asyms; std::map<std::string, std::vector<std::tuple<std::string, int>>> asyms;
...@@ -617,7 +617,7 @@ void comparePolySeqSchemes(datablock &db) ...@@ -617,7 +617,7 @@ void comparePolySeqSchemes(datablock &db)
auto &ndb_poly_seq_scheme = db["ndb_poly_seq_scheme"]; auto &ndb_poly_seq_scheme = db["ndb_poly_seq_scheme"];
auto &pdbx_poly_seq_scheme = db["pdbx_poly_seq_scheme"]; auto &pdbx_poly_seq_scheme = db["pdbx_poly_seq_scheme"];
// Since often ndb_poly_seq_scheme only contains an id and mon_id field // Since often ndb_poly_seq_scheme only contains an id and mon_id item
// we assume that it should match the accompanying pdbx_poly_seq // we assume that it should match the accompanying pdbx_poly_seq
std::vector<std::string> asym_ids_ndb, asym_ids_pdbx; std::vector<std::string> asym_ids_ndb, asym_ids_pdbx;
...@@ -722,6 +722,7 @@ void reconstruct_pdbx(file &file, std::string_view dictionary) ...@@ -722,6 +722,7 @@ void reconstruct_pdbx(file &file, std::string_view dictionary)
std::vector<std::string> invalidCategories; std::vector<std::string> invalidCategories;
// clean up each category
for (auto &cat : db) for (auto &cat : db)
{ {
try try
...@@ -730,12 +731,12 @@ void reconstruct_pdbx(file &file, std::string_view dictionary) ...@@ -730,12 +731,12 @@ void reconstruct_pdbx(file &file, std::string_view dictionary)
if (not cv) if (not cv)
continue; continue;
// Start by renaming columns that may have old names based on alias info // Start by renaming items that may have old names based on alias info
for (auto tag : cat.get_columns()) for (auto tag : cat.get_items())
{ {
auto iv = cv->get_validator_for_item(tag); auto iv = cv->get_validator_for_item(tag);
if (iv) // know, must be OK then if (iv) // know, must be OK then`
continue; continue;
iv = cv->get_validator_for_aliased_item(tag); iv = cv->get_validator_for_aliased_item(tag);
...@@ -744,7 +745,7 @@ void reconstruct_pdbx(file &file, std::string_view dictionary) ...@@ -744,7 +745,7 @@ void reconstruct_pdbx(file &file, std::string_view dictionary)
if (cif::VERBOSE > 0) if (cif::VERBOSE > 0)
std::clog << "Renaming " << tag << " to " << iv->m_tag << " in category " << cat.name() << '\n'; std::clog << "Renaming " << tag << " to " << iv->m_tag << " in category " << cat.name() << '\n';
cat.rename_column(tag, iv->m_tag); cat.rename_item(tag, iv->m_tag);
} }
for (auto link : validator.get_links_for_child(cat.name())) for (auto link : validator.get_links_for_child(cat.name()))
...@@ -767,14 +768,38 @@ void reconstruct_pdbx(file &file, std::string_view dictionary) ...@@ -767,14 +768,38 @@ void reconstruct_pdbx(file &file, std::string_view dictionary)
} }
} }
// Fill in all mandatory fields // Fill in all mandatory items
for (auto key : cv->m_mandatory_fields) for (auto key : cv->m_mandatory_items)
{ {
if (not cat.has_column(key)) if (not cat.has_item(key))
{ {
if (cif::VERBOSE > 0) if (cif::VERBOSE > 0)
std::clog << "Adding mandatory key " << key << " to category " << cat.name() << '\n'; std::clog << "Adding mandatory key " << key << " to category " << cat.name() << '\n';
cat.add_column(key); cat.add_item(key);
}
}
// validate all values, and if they do not validate replace the content with an unknown flag
for (auto tag : cat.get_items())
{
auto iv = cv->get_validator_for_item(tag);
if (not iv)
continue;
auto ix = cat.get_item_ix(tag);
for (auto row : cat)
{
std::error_code ec;
std::string_view value = row[ix].text();
if (not iv->validate_value(value, ec))
{
if (cif::VERBOSE > 0)
std::clog << "Replacing value (" << std::quoted(value) << ") for item " << tag << " since it does not validate\n";
row[ix] = "?";
}
} }
} }
...@@ -834,7 +859,7 @@ void reconstruct_pdbx(file &file, std::string_view dictionary) ...@@ -834,7 +859,7 @@ void reconstruct_pdbx(file &file, std::string_view dictionary)
if (cif::VERBOSE > 0) if (cif::VERBOSE > 0)
std::clog << "Attempt to fix " << cat.name() << " failed: " << ex.what() << '\n'; std::clog << "Attempt to fix " << cat.name() << " failed: " << ex.what() << '\n';
// replace fields that do not define a relation to a parent // replace items that do not define a relation to a parent
std::set<std::string> replaceableKeys; std::set<std::string> replaceableKeys;
for (auto key : cv->m_keys) for (auto key : cv->m_keys)
......
src/pdb/validate-pdbx.cpp
...@@ -69,26 +69,67 @@ condition get_parents_condition(const validator &validator, row_handle rh, const ...@@ -69,26 +69,67 @@ condition get_parents_condition(const validator &validator, row_handle rh, const
bool is_valid_pdbx_file(const file &file, std::string_view dictionary) bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
{ {
using namespace cif::literals; std::error_code ec;
bool result = is_valid_pdbx_file(file, dictionary, ec);
if (ec != std::errc())
throw std::system_error(ec);
return result;
}
auto &cf = cif::compound_factory::instance(); bool is_valid_pdbx_file(const file &file, std::error_code &ec)
auto &validator = cif::validator_factory::instance().operator[](dictionary); {
bool result = false;
if (file.empty())
ec = make_error_code(validation_error::empty_file);
else
{
std::string dictionary = "mmcif_pdbx";
for (auto &db : file)
{
auto audit_conform = db.get("audit_conform");
if (audit_conform == nullptr)
continue;
if (not audit_conform->empty())
{
auto specified_dict = audit_conform->front()["dict_name"];
if (not specified_dict.empty())
dictionary = specified_dict.as<std::string>();
}
break;
}
result = is_valid_pdbx_file(file, dictionary, ec);
}
return result;
}
bool is_valid_pdbx_file(const file &file, std::string_view dictionary, std::error_code &ec)
{
using namespace cif::literals;
bool result = true; bool result = true;
try try
{ {
auto &cf = cif::compound_factory::instance();
auto &validator = cif::validator_factory::instance().operator[](dictionary);
if (file.empty()) if (file.empty())
throw validation_error("Empty file"); throw std::runtime_error("Empty file");
auto &db = file.front(); auto &db = file.front();
if (db.empty()) if (db.empty())
throw validation_error("Empty datablock"); throw std::runtime_error("Empty datablock");
auto &atom_site = db["atom_site"]; auto &atom_site = db["atom_site"];
if (atom_site.empty()) if (atom_site.empty())
throw validation_error("Empty or missing atom_site category"); throw std::runtime_error("Empty or missing atom_site category");
auto &pdbx_poly_seq_scheme = db["pdbx_poly_seq_scheme"]; auto &pdbx_poly_seq_scheme = db["pdbx_poly_seq_scheme"];
...@@ -111,29 +152,29 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -111,29 +152,29 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
auto p = pdbx_poly_seq_scheme.find(get_parents_condition(validator, r, pdbx_poly_seq_scheme)); auto p = pdbx_poly_seq_scheme.find(get_parents_condition(validator, r, pdbx_poly_seq_scheme));
if (p.size() != 1) if (p.size() != 1)
throw validation_error("For each residue in atom_site that is a residue in a polymer there should be exactly one pdbx_poly_seq_scheme record"); throw std::runtime_error("For each residue in atom_site that is a residue in a polymer there should be exactly one pdbx_poly_seq_scheme record");
} }
auto &entity = db["entity"]; auto &entity = db["entity"];
if (entity.empty()) if (entity.empty())
throw validation_error("Entity category is missing or empty"); throw std::runtime_error("Entity category is missing or empty");
auto &entity_poly = db["entity_poly"]; auto &entity_poly = db["entity_poly"];
if (entity_poly.empty()) if (entity_poly.empty())
throw validation_error("Entity_poly category is missing or empty"); throw std::runtime_error("Entity_poly category is missing or empty");
auto &entity_poly_seq = db["entity_poly_seq"]; auto &entity_poly_seq = db["entity_poly_seq"];
if (entity_poly_seq.empty()) if (entity_poly_seq.empty())
throw validation_error("Entity_poly_seq category is missing or empty"); throw std::runtime_error("Entity_poly_seq category is missing or empty");
auto &struct_asym = db["struct_asym"]; auto &struct_asym = db["struct_asym"];
if (struct_asym.empty()) if (struct_asym.empty())
throw validation_error("struct_asym category is missing or empty"); throw std::runtime_error("struct_asym category is missing or empty");
for (auto entity_id : entity.find<std::string>("type"_key == "polymer", "id")) for (auto entity_id : entity.find<std::string>("type"_key == "polymer", "id"))
{ {
if (entity_poly.count("entity_id"_key == entity_id) != 1) if (entity_poly.count("entity_id"_key == entity_id) != 1)
throw validation_error("There should be exactly one entity_poly record per polymer entity"); throw std::runtime_error("There should be exactly one entity_poly record per polymer entity");
const auto entity_poly_type = entity_poly.find1<std::string>("entity_id"_key == entity_id, "type"); const auto entity_poly_type = entity_poly.find1<std::string>("entity_id"_key == entity_id, "type");
...@@ -151,7 +192,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -151,7 +192,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
"seq_id"_key == num and "seq_id"_key == num and
"hetero"_key == hetero) != 1) "hetero"_key == hetero) != 1)
{ {
throw validation_error("For each entity_poly_seq record there should be exactly one pdbx_poly_seq record"); throw std::runtime_error("For each entity_poly_seq record there should be exactly one pdbx_poly_seq record");
} }
} }
} }
...@@ -163,11 +204,11 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -163,11 +204,11 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
"num"_key == seq_id and "num"_key == seq_id and
"hetero"_key == hetero) != 1) "hetero"_key == hetero) != 1)
{ {
throw validation_error("For each pdbx_poly_seq/struct_asym record there should be exactly one entity_poly_seq record"); throw std::runtime_error("For each pdbx_poly_seq/struct_asym record there should be exactly one entity_poly_seq record");
} }
if ((mon_per_seq_id[seq_id].size() > 1) != hetero) if ((mon_per_seq_id[seq_id].size() > 1) != hetero)
throw validation_error("Mismatch between the hetero flag in the poly seq schemes and the number residues per seq_id"); throw std::runtime_error("Mismatch between the hetero flag in the poly seq schemes and the number residues per seq_id");
} }
for (const auto &[seq_id, mon_ids] : mon_per_seq_id) for (const auto &[seq_id, mon_ids] : mon_per_seq_id)
...@@ -184,7 +225,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -184,7 +225,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
"label_seq_id"_key == seq_id and not std::move(cond); "label_seq_id"_key == seq_id and not std::move(cond);
if (atom_site.contains(std::move(cond))) if (atom_site.contains(std::move(cond)))
throw validation_error("An atom_site record exists that has no parent in the poly seq scheme categories"); throw std::runtime_error("An atom_site record exists that has no parent in the poly seq scheme categories");
} }
} }
...@@ -250,7 +291,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -250,7 +291,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
seq->erase(std::remove_if(seq->begin(), seq->end(), [](char ch) { return std::isspace(ch); }), seq->end()); seq->erase(std::remove_if(seq->begin(), seq->end(), [](char ch) { return std::isspace(ch); }), seq->end());
if (not seq_match(false, seq->begin(), seq->end())) if (not seq_match(false, seq->begin(), seq->end()))
throw validation_error("Sequences do not match for entity " + entity_id); throw std::runtime_error("Sequences do not match for entity " + entity_id);
} }
if (not seq_can.has_value()) if (not seq_can.has_value())
...@@ -263,7 +304,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -263,7 +304,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
seq_can->erase(std::remove_if(seq_can->begin(), seq_can->end(), [](char ch) { return std::isspace(ch); }), seq_can->end()); seq_can->erase(std::remove_if(seq_can->begin(), seq_can->end(), [](char ch) { return std::isspace(ch); }), seq_can->end());
if (not seq_match(true, seq_can->begin(), seq_can->end())) if (not seq_match(true, seq_can->begin(), seq_can->end()))
throw validation_error("Canonical sequences do not match for entity " + entity_id); throw std::runtime_error("Canonical sequences do not match for entity " + entity_id);
} }
} }
...@@ -275,6 +316,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary) ...@@ -275,6 +316,7 @@ bool is_valid_pdbx_file(const file &file, std::string_view dictionary)
result = false; result = false;
if (cif::VERBOSE > 0) if (cif::VERBOSE > 0)
std::clog << ex.what() << '\n'; std::clog << ex.what() << '\n';
ec = make_error_code(validation_error::not_valid_pdbx);
} }
return result; return result;
......
src/row.cpp
...@@ -29,44 +29,44 @@ ...@@ -29,44 +29,44 @@
namespace cif namespace cif
{ {
void row_handle::assign(uint16_t column, std::string_view value, bool updateLinked, bool validate) void row_handle::assign(uint16_t item, std::string_view value, bool updateLinked, bool validate)
{ {
if (not m_category) if (not m_category)
throw std::runtime_error("uninitialized row"); throw std::runtime_error("uninitialized row");
m_category->update_value(m_row, column, value, updateLinked, validate); m_category->update_value(m_row, item, value, updateLinked, validate);
} }
uint16_t row_handle::get_column_ix(std::string_view name) const uint16_t row_handle::get_item_ix(std::string_view name) const
{ {
if (not m_category) if (not m_category)
throw std::runtime_error("uninitialized row"); throw std::runtime_error("uninitialized row");
return m_category->get_column_ix(name); return m_category->get_item_ix(name);
} }
std::string_view row_handle::get_column_name(uint16_t ix) const std::string_view row_handle::get_item_name(uint16_t ix) const
{ {
if (not m_category) if (not m_category)
throw std::runtime_error("uninitialized row"); throw std::runtime_error("uninitialized row");
return m_category->get_column_name(ix); return m_category->get_item_name(ix);
} }
uint16_t row_handle::add_column(std::string_view name) uint16_t row_handle::add_item(std::string_view name)
{ {
if (not m_category) if (not m_category)
throw std::runtime_error("uninitialized row"); throw std::runtime_error("uninitialized row");
return m_category->add_column(name); return m_category->add_item(name);
} }
void row_handle::swap(uint16_t column, row_handle &b) void row_handle::swap(uint16_t item, row_handle &b)
{ {
if (not m_category) if (not m_category)
throw std::runtime_error("uninitialized row"); throw std::runtime_error("uninitialized row");
m_category->swap_item(column, *this, b); m_category->swap_item(item, *this, b);
} }
// -------------------------------------------------------------------- // --------------------------------------------------------------------
...@@ -86,7 +86,7 @@ row_initializer::row_initializer(row_handle rh) ...@@ -86,7 +86,7 @@ row_initializer::row_initializer(row_handle rh)
auto &i = r->operator[](ix); auto &i = r->operator[](ix);
if (not i) if (not i)
continue; continue;
emplace_back(cat.get_column_name(ix), i.text()); emplace_back(cat.get_item_name(ix), i.text());
} }
} }
......
src/validate.cpp
...@@ -39,10 +39,10 @@ ...@@ -39,10 +39,10 @@
// the code will use boost::regex instead. // the code will use boost::regex instead.
#if USE_BOOST_REGEX #if USE_BOOST_REGEX
#include <boost/regex.hpp> # include <boost/regex.hpp>
using boost::regex; using boost::regex;
#else #else
#include <regex> # include <regex>
using std::regex; using std::regex;
#endif #endif
...@@ -57,20 +57,11 @@ struct regex_impl : public regex ...@@ -57,20 +57,11 @@ struct regex_impl : public regex
} }
}; };
validation_error::validation_error(const std::string &msg)
: m_msg(msg)
{
}
validation_error::validation_error(const std::string &cat, const std::string &item, const std::string &msg)
: m_msg("When validating _" + cat + '.' + item + ": " + msg)
{
}
// -------------------------------------------------------------------- // --------------------------------------------------------------------
DDL_PrimitiveType map_to_primitive_type(std::string_view s) DDL_PrimitiveType map_to_primitive_type(std::string_view s, std::error_code &ec) noexcept
{ {
ec = {};
DDL_PrimitiveType result; DDL_PrimitiveType result;
if (iequals(s, "char")) if (iequals(s, "char"))
result = DDL_PrimitiveType::Char; result = DDL_PrimitiveType::Char;
...@@ -79,7 +70,16 @@ DDL_PrimitiveType map_to_primitive_type(std::string_view s) ...@@ -79,7 +70,16 @@ DDL_PrimitiveType map_to_primitive_type(std::string_view s)
else if (iequals(s, "numb")) else if (iequals(s, "numb"))
result = DDL_PrimitiveType::Numb; result = DDL_PrimitiveType::Numb;
else else
throw validation_error("Not a known primitive type"); ec = make_error_code(validation_error::not_a_known_primitive_type);
return result;
}
DDL_PrimitiveType map_to_primitive_type(std::string_view s)
{
std::error_code ec;
auto result = map_to_primitive_type(s, ec);
if (ec)
throw std::system_error(ec, std::string{ s });
return result; return result;
} }
...@@ -218,17 +218,24 @@ int type_validator::compare(std::string_view a, std::string_view b) const ...@@ -218,17 +218,24 @@ int type_validator::compare(std::string_view a, std::string_view b) const
void item_validator::operator()(std::string_view value) const void item_validator::operator()(std::string_view value) const
{ {
std::error_code ec;
if (not validate_value(value, ec))
throw std::system_error(ec, std::string{ value } + " does not match rx for " + m_tag);
}
bool item_validator::validate_value(std::string_view value, std::error_code &ec) const noexcept
{
ec = {};
if (not value.empty() and value != "?" and value != ".") if (not value.empty() and value != "?" and value != ".")
{ {
if (m_type != nullptr and not regex_match(value.begin(), value.end(), *m_type->m_rx)) if (m_type != nullptr and not regex_match(value.begin(), value.end(), *m_type->m_rx))
throw validation_error(m_category->m_name, m_tag, "Value '" + std::string{ value } + "' does not match type expression for type " + m_type->m_name); ec = make_error_code(validation_error::value_does_not_match_rx);
else if (not m_enums.empty() and m_enums.count(std::string{ value }) == 0)
if (not m_enums.empty()) ec = make_error_code(validation_error::value_is_not_in_enumeration_list);
{
if (m_enums.count(std::string{ value }) == 0)
throw validation_error(m_category->m_name, m_tag, "Value '" + std::string{ value } + "' is not in the list of allowed values");
}
} }
return ec == std::errc();
} }
// -------------------------------------------------------------------- // --------------------------------------------------------------------
...@@ -236,7 +243,7 @@ void item_validator::operator()(std::string_view value) const ...@@ -236,7 +243,7 @@ void item_validator::operator()(std::string_view value) const
void category_validator::add_item_validator(item_validator &&v) void category_validator::add_item_validator(item_validator &&v)
{ {
if (v.m_mandatory) if (v.m_mandatory)
m_mandatory_fields.insert(v.m_tag); m_mandatory_items.insert(v.m_tag);
v.m_category = this; v.m_category = this;
...@@ -265,7 +272,7 @@ const item_validator *category_validator::get_validator_for_aliased_item(std::st ...@@ -265,7 +272,7 @@ const item_validator *category_validator::get_validator_for_aliased_item(std::st
for (auto &ai : iv.m_aliases) for (auto &ai : iv.m_aliases)
{ {
const auto &[cat, name] = split_tag_name(ai.m_name); const auto &[cat, name] = split_tag_name(ai.m_name);
if (name == tag and cat == m_name) if (iequals(name, tag) and iequals(cat, m_name))
{ {
result = &iv; result = &iv;
break; break;
...@@ -373,7 +380,7 @@ std::vector<const link_validator *> validator::get_links_for_parent(std::string_ ...@@ -373,7 +380,7 @@ std::vector<const link_validator *> validator::get_links_for_parent(std::string_
for (auto &l : m_link_validators) for (auto &l : m_link_validators)
{ {
if (l.m_parent_category == category) if (iequals(l.m_parent_category, category))
result.push_back(&l); result.push_back(&l);
} }
...@@ -386,19 +393,41 @@ std::vector<const link_validator *> validator::get_links_for_child(std::string_v ...@@ -386,19 +393,41 @@ std::vector<const link_validator *> validator::get_links_for_child(std::string_v
for (auto &l : m_link_validators) for (auto &l : m_link_validators)
{ {
if (l.m_child_category == category) if (iequals(l.m_child_category, category))
result.push_back(&l); result.push_back(&l);
} }
return result; return result;
} }
void validator::report_error(const std::string &msg, bool fatal) const // void validator::report_error(const std::string &msg, bool fatal) const
// {
// if (m_strict or fatal)
// throw validation_error(msg);
// else if (VERBOSE > 0)
// std::cerr << msg << '\n';
// }
void validator::report_error(std::error_code ec, bool fatal) const
{ {
if (m_strict or fatal) if (m_strict or fatal)
throw validation_error(msg); throw std::system_error(ec);
else if (VERBOSE > 0) else
std::cerr << msg << '\n'; std::cerr << ec.message() << '\n';
}
void validator::report_error(std::error_code ec, std::string_view category,
std::string_view item, bool fatal) const
{
std::ostringstream os;
os << "category: "<< category;
if (not item.empty())
os << "; item: " << item;
if (m_strict or fatal)
throw std::system_error(ec, os.str());
else
std::cerr << ec.message() << ": " << os.str() << '\n';
} }
// -------------------------------------------------------------------- // --------------------------------------------------------------------
...@@ -460,12 +489,12 @@ const validator &validator_factory::operator[](std::string_view dictionary_name) ...@@ -460,12 +489,12 @@ const validator &validator_factory::operator[](std::string_view dictionary_name)
if (not std::filesystem::exists(p, ec) or ec) if (not std::filesystem::exists(p, ec) or ec)
{ {
for (const char *dir : { for (const char *dir : {
#if defined(CACHE_DIR) # if defined(CACHE_DIR)
CACHE_DIR, CACHE_DIR,
#endif # endif
#if defined(DATA_DIR) # if defined(DATA_DIR)
DATA_DIR DATA_DIR
#endif # endif
}) })
{ {
auto p2 = std::filesystem::path(dir) / p; auto p2 = std::filesystem::path(dir) / p;
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment